mice
2025-01-28
Paqueteria necesaria
library(tableone)
library(Amelia)
library(naniar)
library(tidyr)
library(ggplot2)
library(mice)
library(readxl)PASO I: Carga de Datos
library("dplyr")
data <- read_excel("/Users/angelaluquegarcia/Documents/Diseño_Bioinfo/Base_de_datos_IAM.xlsx")
summary(data)## Id NivelSocieconomico Age Sex Admission_Date
## Min. : 1.00 Min. : 0.00 Min. :34.00 Min. :0.0000 Min. :2010-01-24 00:00:00.00
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 1st Qu.:0.0000 1st Qu.:2012-05-11 12:00:00.00
## Median :160.50 Median :157.50 Median :56.00 Median :1.0000 Median :2015-07-26 00:00:00.00
## Mean :160.50 Mean :158.31 Mean :54.98 Mean :0.6773 Mean :2015-07-09 12:07:07.72
## 3rd Qu.:240.25 3rd Qu.:238.25 3rd Qu.:62.00 3rd Qu.:1.0000 3rd Qu.:2018-08-21 00:00:00.00
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :1.0000 Max. :2020-12-16 00:00:00.00
## NA's :1 NA's :13 NA's :13 NA's :8 NA's :18
## Chest pain type HTA FA BP Cholesterol FBS over 120
## Min. :1.000 Min. :0.0000 Min. :0.0000 Min. : 94.0 Min. :126.0 Min. :0.0000
## 1st Qu.:3.000 1st Qu.:0.0000 1st Qu.:0.0000 1st Qu.:120.0 1st Qu.:214.0 1st Qu.:0.0000
## Median :3.000 Median :0.0000 Median :0.0000 Median :130.0 Median :244.5 Median :0.0000
## Mean :3.173 Mean :0.2437 Mean :0.2461 Mean :130.8 Mean :249.2 Mean :0.1373
## 3rd Qu.:4.000 3rd Qu.:0.0000 3rd Qu.:0.0000 3rd Qu.:140.0 3rd Qu.:277.0 3rd Qu.:0.0000
## Max. :4.000 Max. :1.0000 Max. :1.0000 Max. :192.0 Max. :417.0 Max. :1.0000
## NA's :20 NA's :1 NA's :18 NA's :19 NA's :15
## EKG results Max HR Exercise angina ST depression Slope of ST Number of vessels fluro
## Min. :0.00 Min. : 71.0 Min. :0.0000 Min. :0.000 Min. :1.000 Min. :0.0000
## 1st Qu.:0.00 1st Qu.:137.8 1st Qu.:0.0000 1st Qu.:0.000 1st Qu.:1.000 1st Qu.:0.0000
## Median :1.50 Median :153.0 Median :0.0000 Median :0.800 Median :2.000 Median :0.0000
## Mean :1.01 Mean :149.6 Mean :0.3322 Mean :1.091 Mean :1.594 Mean :0.6964
## 3rd Qu.:2.00 3rd Qu.:165.0 3rd Qu.:1.0000 3rd Qu.:1.800 3rd Qu.:2.000 3rd Qu.:1.0000
## Max. :2.00 Max. :202.0 Max. :1.0000 Max. :6.200 Max. :3.000 Max. :3.0000
## NA's :23 NA's :17 NA's :20 NA's :13 NA's :18 NA's :18
## Thallium BiomarcadorBasal_A BiomarcadorBasal_B Heart Disease Cholesterol_Level Glucose_Level
## Min. :3.000 Min. : 1.636 Min. : 1.565 Min. :0.0000 Min. : 69.01 Min. : 65.47
## 1st Qu.:3.000 1st Qu.: 7.537 1st Qu.: 5.499 1st Qu.:0.0000 1st Qu.:170.50 1st Qu.: 87.14
## Median :3.000 Median : 9.650 Median : 6.889 Median :0.0000 Median :205.26 Median : 99.34
## Mean :4.757 Mean : 11.096 Mean : 7.976 Mean :0.4488 Mean :204.40 Mean : 99.02
## 3rd Qu.:7.000 3rd Qu.: 11.750 3rd Qu.: 8.251 3rd Qu.:1.0000 3rd Qu.:234.01 3rd Qu.:108.45
## Max. :7.000 Max. :435.045 Max. :350.024 Max. :1.0000 Max. :392.64 Max. :146.18
## NA's :17 NA's :12 NA's :14 NA's :18 NA's :185 NA's :185
## Neutrophils_Count Platelets_Count LDL_Colesterol BiomarcadorAgudo_C Symptom_Scale_Admission
## Min. :1.890 Min. : 45.62 Min. : 0.00 Min. : 23.92 Min. : 0.00
## 1st Qu.:4.199 1st Qu.:209.48 1st Qu.: 88.64 1st Qu.: 46.37 1st Qu.: 8.00
## Median :5.033 Median :247.68 Median :143.02 Median : 60.97 Median :19.00
## Mean :5.118 Mean :251.91 Mean :144.78 Mean : 67.53 Mean :20.53
## 3rd Qu.:6.015 3rd Qu.:292.82 3rd Qu.:205.73 3rd Qu.: 79.83 3rd Qu.:34.00
## Max. :8.949 Max. :404.78 Max. :306.48 Max. :218.84 Max. :48.00
## NA's :185 NA's :185 NA's :192 NA's :164 NA's :184
## Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge Treatment_Response Treatment_Type
## Min. : 0.00 Length:321 Length:321 Length:321 Length:321
## 1st Qu.:16.00 Class :character Class :character Class :character Class :character
## Median :26.00 Mode :character Mode :character Mode :character Mode :character
## Mean :26.31
## 3rd Qu.:40.00
## Max. :50.00
## NA's :192
## 3M_Prognosis Death_Date
## Length:321 Min. :2010-11-23 00:00:00.00
## Class :character 1st Qu.:2015-04-27 00:00:00.00
## Mode :character Median :2019-06-24 00:00:00.00
## Mean :2018-01-19 01:48:35.41
## 3rd Qu.:2019-12-30 00:00:00.00
## Max. :2023-11-27 00:00:00.00
## NA's :16se realiza un vector con el título de las columnas de interés, estas serán las mostradas en nuestro diccionario como datos antes.
selected_columns <- c("Id",
"NivelSocieconomico",
"Age",
"Sex",
"Admission_Date",
"Chest pain type",
"BP",
"Cholesterol",
"FBS over 120",
"EKG results",
"Max HR",
"Exercise angina",
"ST depression",
"Slope of ST",
"Number of vessels fluro",
"Thallium",
"BiomarcadorBasal_A",
"BiomarcadorBasal_B",
"Heart Disease")
data_antes <- data[, selected_columns]Paso II:
Observamos los datatos cuyo porcentaje de misvalues es menor al 20% pueden ser imputados. La columna ID parece tener solo una valor nan cuyas variables tambien lo son, por lo que eliminamos esa la fila completa Damos un repaso a nuestra base de datos para revisar que loa media de hoy esta en orden y
data_antes <- data_antes %>% filter(!is.na(Id)) # Eliminacion de fila
summary(data_antes) #verificacion de la dat## Id NivelSocieconomico Age Sex Admission_Date
## Min. : 1.00 Min. : 0.00 Min. :34.00 Min. :0.0000 Min. :2010-01-24 00:00:00.00
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 1st Qu.:0.0000 1st Qu.:2012-05-11 12:00:00.00
## Median :160.50 Median :157.50 Median :56.00 Median :1.0000 Median :2015-07-26 00:00:00.00
## Mean :160.50 Mean :158.31 Mean :54.98 Mean :0.6773 Mean :2015-07-09 12:07:07.72
## 3rd Qu.:240.25 3rd Qu.:238.25 3rd Qu.:62.00 3rd Qu.:1.0000 3rd Qu.:2018-08-21 00:00:00.00
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :1.0000 Max. :2020-12-16 00:00:00.00
## NA's :12 NA's :12 NA's :7 NA's :17
## Chest pain type BP Cholesterol FBS over 120 EKG results Max HR
## Min. :1.000 Min. : 94.0 Min. :126.0 Min. :0.0000 Min. :0.00 Min. : 71.0
## 1st Qu.:3.000 1st Qu.:120.0 1st Qu.:214.0 1st Qu.:0.0000 1st Qu.:0.00 1st Qu.:137.8
## Median :3.000 Median :130.0 Median :244.5 Median :0.0000 Median :1.50 Median :153.0
## Mean :3.173 Mean :130.8 Mean :249.2 Mean :0.1373 Mean :1.01 Mean :149.6
## 3rd Qu.:4.000 3rd Qu.:140.0 3rd Qu.:277.0 3rd Qu.:0.0000 3rd Qu.:2.00 3rd Qu.:165.0
## Max. :4.000 Max. :192.0 Max. :417.0 Max. :1.0000 Max. :2.00 Max. :202.0
## NA's :19 NA's :17 NA's :18 NA's :14 NA's :22 NA's :16
## Exercise angina ST depression Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A
## Min. :0.0000 Min. :0.000 Min. :1.000 Min. :0.0000 Min. :3.000 Min. : 1.636
## 1st Qu.:0.0000 1st Qu.:0.000 1st Qu.:1.000 1st Qu.:0.0000 1st Qu.:3.000 1st Qu.: 7.537
## Median :0.0000 Median :0.800 Median :2.000 Median :0.0000 Median :3.000 Median : 9.650
## Mean :0.3322 Mean :1.091 Mean :1.594 Mean :0.6964 Mean :4.757 Mean : 11.096
## 3rd Qu.:1.0000 3rd Qu.:1.800 3rd Qu.:2.000 3rd Qu.:1.0000 3rd Qu.:7.000 3rd Qu.: 11.750
## Max. :1.0000 Max. :6.200 Max. :3.000 Max. :3.0000 Max. :7.000 Max. :435.045
## NA's :19 NA's :12 NA's :17 NA's :17 NA's :16 NA's :11
## BiomarcadorBasal_B Heart Disease
## Min. : 1.565 Min. :0.0000
## 1st Qu.: 5.499 1st Qu.:0.0000
## Median : 6.889 Median :0.0000
## Mean : 7.976 Mean :0.4488
## 3rd Qu.: 8.251 3rd Qu.:1.0000
## Max. :350.024 Max. :1.0000
## NA's :13 NA's :17Se hace un conteo de miss valué en la base de datos para sacar su porcentaje y saber qué tipo de imputación es la más adecuada. Al no haber porcentajes mayores al 20% estos pueden ser imputados por media, moda o predicción.
#get percentage of missing value of the attributes - Approach 2 (Function)
sapply(data_antes, function(df)
{
sum(is.na(df)==T)/length(df)*100
})## Id NivelSocieconomico Age Sex
## 0.0000 3.7500 3.7500 2.1875
## Admission_Date Chest pain type BP Cholesterol
## 5.3125 5.9375 5.3125 5.6250
## FBS over 120 EKG results Max HR Exercise angina
## 4.3750 6.8750 5.0000 5.9375
## ST depression Slope of ST Number of vessels fluro Thallium
## 3.7500 5.3125 5.3125 5.0000
## BiomarcadorBasal_A BiomarcadorBasal_B Heart Disease
## 3.4375 4.0625 5.3125missmap(data_antes, main = "Missing Map", col = c("yellow", "black"), legend = TRUE)## Warning: Unknown or uninitialised column: `arguments`.
## Unknown or uninitialised column: `arguments`.## Warning: Unknown or uninitialised column: `imputations`.
# Visualización de valores faltantes con naniar
vis_miss(data_antes) +
labs(title = "Mapa de valores faltantes")
# Porcentaje de valores faltantes
missing_summary <- data_antes %>%
summarise(across(everything(), ~ mean(is.na(.)) * 100)) %>%
pivot_longer(cols = everything(), names_to = "Variable", values_to = "Missing_Percentage")
# Visualización del porcentaje de valores faltantes
ggplot(missing_summary, aes(x = reorder(Variable, -Missing_Percentage), y = Missing_Percentage)) +
geom_col(fill = "red") +
coord_flip() +
labs(title = "Porcentaje de valores faltantes por variable", x = "Variable", y = "Porcentaje")
### *** Paso III: Procesamiento de los datos he imputacion.****
Para tener mejores resultados en la imputación nos aseguramos que las variables categóricas estén clasificadas como un factor.
# Convertir la columna 'Gender' a factor con los niveles 'Masculino' y 'Femenino'
data_antes$Sex <- as.factor(data_antes$Sex)
levels(data_antes$Sex) <- c("Femenino", "Masculino")
data_antes <- data_antes %>% mutate(across(where(~ all(. %in% c(0, 1))), ~ ifelse(. == 1, "SI", "NO")))
# Convertir columnas no numéricas a factores para imputación
factor_columns <- sapply(data_antes, is.character) | sapply(data_antes, is.logical)
data_antes[factor_columns] <- lapply(data_antes[factor_columns], as.factor)
# 4. Configurar la matriz de predicción (opcional)
pred_matrix <- quickpred(data_antes, mincor = 0.1, minpuc = 0.5)
# categorizar columanas segun el diccionario.
# Convertir columnas específicas en factores si existen
columns_to_factor <- c("Chest pain type", "FBS over 120", "EKG results", "Exercise angina",
"Slope of ST", "Thallium", "Heart Disease", "Killip_Class_24H",
"Recovery_Discharge", "Treatment_Response", "Treatment_Type",
"3M_Prognosis")
existing_columns <- columns_to_factor[columns_to_factor %in% names(data_antes)]
data_antes[existing_columns] <- lapply(data_antes[existing_columns], as.factor)
data_antes$`FBS over 120` <- as.factor(ifelse(data_antes$`FBS over 120` == 1, "Si", "No"))
data_antes$`Exercise angina` <- as.factor(ifelse(data_antes$`Exercise angina` == 1, "Si", "No"))
data_antes$`Heart Disease` <- as.factor(ifelse(data_antes$`Heart Disease` == 1, "Presence", "Absence"))
lapply(data_antes, class)## $Id
## [1] "numeric"
##
## $NivelSocieconomico
## [1] "numeric"
##
## $Age
## [1] "numeric"
##
## $Sex
## [1] "factor"
##
## $Admission_Date
## [1] "POSIXct" "POSIXt"
##
## $`Chest pain type`
## [1] "factor"
##
## $BP
## [1] "numeric"
##
## $Cholesterol
## [1] "numeric"
##
## $`FBS over 120`
## [1] "factor"
##
## $`EKG results`
## [1] "factor"
##
## $`Max HR`
## [1] "numeric"
##
## $`Exercise angina`
## [1] "factor"
##
## $`ST depression`
## [1] "numeric"
##
## $`Slope of ST`
## [1] "factor"
##
## $`Number of vessels fluro`
## [1] "numeric"
##
## $Thallium
## [1] "factor"
##
## $BiomarcadorBasal_A
## [1] "numeric"
##
## $BiomarcadorBasal_B
## [1] "numeric"
##
## $`Heart Disease`
## [1] "factor"summary(data_antes)## Id NivelSocieconomico Age Sex Admission_Date
## Min. : 1.00 Min. : 0.00 Min. :34.00 Femenino :101 Min. :2010-01-24 00:00:00.00
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 Masculino:212 1st Qu.:2012-05-11 12:00:00.00
## Median :160.50 Median :157.50 Median :56.00 NA's : 7 Median :2015-07-26 00:00:00.00
## Mean :160.50 Mean :158.31 Mean :54.98 Mean :2015-07-09 12:07:07.72
## 3rd Qu.:240.25 3rd Qu.:238.25 3rd Qu.:62.00 3rd Qu.:2018-08-21 00:00:00.00
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :2020-12-16 00:00:00.00
## NA's :12 NA's :12 NA's :17
## Chest pain type BP Cholesterol FBS over 120 EKG results Max HR Exercise angina
## 1 : 21 Min. : 94.0 Min. :126.0 No :264 0 :146 Min. : 71.0 No :201
## 2 : 46 1st Qu.:120.0 1st Qu.:214.0 Si : 42 1 : 3 1st Qu.:137.8 Si :100
## 3 : 94 Median :130.0 Median :244.5 NA's: 14 2 :149 Median :153.0 NA's: 19
## 4 :140 Mean :130.8 Mean :249.2 NA's: 22 Mean :149.6
## NA's: 19 3rd Qu.:140.0 3rd Qu.:277.0 3rd Qu.:165.0
## Max. :192.0 Max. :417.0 Max. :202.0
## NA's :17 NA's :18 NA's :16
## ST depression Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B
## Min. :0.000 1 :143 Min. :0.0000 3 :166 Min. : 1.636 Min. : 1.565
## 1st Qu.:0.000 2 :140 1st Qu.:0.0000 6 : 18 1st Qu.: 7.537 1st Qu.: 5.499
## Median :0.800 3 : 20 Median :0.0000 7 :120 Median : 9.650 Median : 6.889
## Mean :1.091 NA's: 17 Mean :0.6964 NA's: 16 Mean : 11.096 Mean : 7.976
## 3rd Qu.:1.800 3rd Qu.:1.0000 3rd Qu.: 11.750 3rd Qu.: 8.251
## Max. :6.200 Max. :3.0000 Max. :435.045 Max. :350.024
## NA's :12 NA's :17 NA's :11 NA's :13
## Heart Disease
## Absence :167
## Presence:136
## NA's : 17
##
##
##
## Para la imputación de variables numéricas se usa la paquetería “mice” con el método “PMM”
numeric_columns <- sapply(data_antes, is.numeric)
numeric_data <- data_antes[, numeric_columns]
# Realizar imputación con PMM (Predictive Mean Matching)
imputed_numeric <- mice(numeric_data, method = "pmm", m = 5, maxit = 50, seed = 123)##
## iter imp variable
## 1 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 1 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 1 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 1 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 1 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 2 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 2 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 2 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 2 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 2 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 3 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 3 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 3 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 3 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 3 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 4 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 4 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 4 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 4 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 4 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 5 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 5 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 5 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 5 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 5 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 6 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 6 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 6 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 6 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 6 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 7 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 7 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 7 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 7 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 7 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 8 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 8 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 8 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 8 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 8 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 9 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 9 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 9 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 9 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 9 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 10 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 10 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 10 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 10 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 10 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 11 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 11 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 11 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 11 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 11 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 12 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 12 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 12 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 12 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 12 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 13 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 13 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 13 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 13 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 13 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 14 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 14 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 14 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 14 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 14 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 15 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 15 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 15 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 15 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 15 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 16 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 16 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 16 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 16 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 16 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 17 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 17 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 17 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 17 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 17 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 18 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 18 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 18 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 18 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 18 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 19 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 19 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 19 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 19 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 19 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 20 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 20 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 20 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 20 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 20 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 21 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 21 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 21 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 21 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 21 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 22 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 22 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 22 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 22 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 22 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 23 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 23 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 23 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 23 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 23 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 24 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 24 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 24 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 24 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 24 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 25 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 25 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 25 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 25 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 25 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 26 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 26 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 26 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 26 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 26 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 27 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 27 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 27 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 27 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 27 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 28 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 28 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 28 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 28 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 28 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 29 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 29 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 29 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 29 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 29 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 30 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 30 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 30 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 30 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 30 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 31 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 31 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 31 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 31 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 31 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 32 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 32 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 32 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 32 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 32 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 33 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 33 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 33 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 33 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 33 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 34 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 34 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 34 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 34 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 34 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 35 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 35 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 35 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 35 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 35 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 36 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 36 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 36 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 36 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 36 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 37 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 37 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 37 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 37 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 37 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 38 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 38 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 38 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 38 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 38 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 39 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 39 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 39 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 39 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 39 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 40 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 40 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 40 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 40 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 40 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 41 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 41 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 41 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 41 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 41 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 42 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 42 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 42 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 42 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 42 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 43 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 43 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 43 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 43 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 43 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 44 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 44 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 44 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 44 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 44 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 45 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 45 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 45 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 45 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 45 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 46 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 46 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 46 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 46 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 46 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 47 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 47 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 47 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 47 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 47 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 48 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 48 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 48 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 48 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 48 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 49 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 49 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 49 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 49 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 49 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 50 1 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 50 2 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 50 3 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 50 4 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_B
## 50 5 NivelSocieconomico Age BP Cholesterol Max HR ST depression Number of vessels fluro BiomarcadorBasal_A BiomarcadorBasal_Bnumeric_data <- complete(imputed_numeric)
data_antes[, numeric_columns] <- numeric_data
summary(data_antes)## Id NivelSocieconomico Age Sex Admission_Date
## Min. : 1.00 Min. : 0.00 Min. :34.00 Femenino :101 Min. :2010-01-24 00:00:00.00
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 Masculino:212 1st Qu.:2012-05-11 12:00:00.00
## Median :160.50 Median :156.50 Median :56.00 NA's : 7 Median :2015-07-26 00:00:00.00
## Mean :160.50 Mean :157.85 Mean :54.99 Mean :2015-07-09 12:07:07.72
## 3rd Qu.:240.25 3rd Qu.:237.25 3rd Qu.:62.00 3rd Qu.:2018-08-21 00:00:00.00
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :2020-12-16 00:00:00.00
## NA's :17
## Chest pain type BP Cholesterol FBS over 120 EKG results Max HR Exercise angina
## 1 : 21 Min. : 94.0 Min. :126.0 No :264 0 :146 Min. : 71.0 No :201
## 2 : 46 1st Qu.:120.0 1st Qu.:212.8 Si : 42 1 : 3 1st Qu.:135.5 Si :100
## 3 : 94 Median :130.0 Median :244.5 NA's: 14 2 :149 Median :152.0 NA's: 19
## 4 :140 Mean :131.3 Mean :248.8 NA's: 22 Mean :149.0
## NA's: 19 3rd Qu.:140.0 3rd Qu.:281.2 3rd Qu.:164.2
## Max. :192.0 Max. :417.0 Max. :202.0
##
## ST depression Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B
## Min. :0.0 1 :143 Min. :0.0000 3 :166 Min. : 1.636 Min. : 1.565
## 1st Qu.:0.0 2 :140 1st Qu.:0.0000 6 : 18 1st Qu.: 7.508 1st Qu.: 5.494
## Median :0.8 3 : 20 Median :0.0000 7 :120 Median : 9.660 Median : 6.918
## Mean :1.1 NA's: 17 Mean :0.7031 NA's: 16 Mean : 11.054 Mean : 7.931
## 3rd Qu.:1.8 3rd Qu.:1.0000 3rd Qu.: 11.751 3rd Qu.: 8.247
## Max. :6.2 Max. :3.0000 Max. :435.045 Max. :350.024
##
## Heart Disease
## Absence :167
## Presence:136
## NA's : 17
##
##
##
## Imputación para variables categóricas ordinales con regrecion logistica “logrer”
# Seleccionar columnas categóricas
categorical_columns <- sapply(data_antes, is.factor) | sapply(data_antes, is.character)
categorical_data <- data_antes[, categorical_columns]
# Convertir columnas de texto a factores (si es necesario)
categorical_data <- lapply(categorical_data, as.factor)
categorical_data <- as.data.frame(categorical_data)
# Realizar imputación con regresión logística (logreg)
imputed_categorical <- mice(categorical_data, method = "cart", m = 5, maxit = 50, seed = 123)##
## iter imp variable
## 1 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 1 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 1 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 1 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 1 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 2 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 2 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 2 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 2 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 2 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 3 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 3 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 3 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 3 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 3 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 4 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 4 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 4 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 4 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 4 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 5 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 5 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 5 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 5 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 5 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 6 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 6 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 6 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 6 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 6 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 7 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 7 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 7 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 7 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 7 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 8 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 8 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 8 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 8 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 8 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 9 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 9 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 9 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 9 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 9 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 10 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 10 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 10 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 10 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 10 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 11 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 11 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 11 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 11 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 11 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 12 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 12 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 12 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 12 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 12 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 13 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 13 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 13 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 13 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 13 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 14 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 14 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 14 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 14 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 14 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 15 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 15 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 15 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 15 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 15 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 16 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 16 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 16 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 16 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 16 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 17 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 17 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 17 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 17 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 17 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 18 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 18 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 18 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 18 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 18 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 19 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 19 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 19 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 19 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 19 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 20 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 20 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 20 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 20 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 20 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 21 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 21 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 21 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 21 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 21 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 22 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 22 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 22 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 22 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 22 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 23 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 23 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 23 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 23 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 23 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 24 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 24 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 24 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 24 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 24 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 25 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 25 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 25 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 25 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 25 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 26 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 26 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 26 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 26 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 26 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 27 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 27 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 27 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 27 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 27 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 28 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 28 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 28 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 28 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 28 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 29 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 29 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 29 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 29 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 29 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 30 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 30 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 30 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 30 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 30 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 31 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 31 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 31 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 31 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 31 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 32 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 32 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 32 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 32 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 32 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 33 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 33 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 33 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 33 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 33 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 34 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 34 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 34 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 34 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 34 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 35 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 35 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 35 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 35 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 35 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 36 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 36 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 36 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 36 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 36 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 37 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 37 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 37 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 37 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 37 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 38 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 38 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 38 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 38 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 38 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 39 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 39 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 39 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 39 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 39 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 40 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 40 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 40 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 40 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 40 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 41 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 41 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 41 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 41 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 41 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 42 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 42 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 42 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 42 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 42 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 43 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 43 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 43 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 43 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 43 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 44 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 44 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 44 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 44 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 44 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 45 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 45 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 45 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 45 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 45 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 46 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 46 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 46 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 46 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 46 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 47 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 47 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 47 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 47 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 47 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 48 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 48 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 48 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 48 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 48 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 49 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 49 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 49 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 49 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 49 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 50 1 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 50 2 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 50 3 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 50 4 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Disease
## 50 5 Sex Chest.pain.type FBS.over.120 EKG.results Exercise.angina Slope.of.ST Thallium Heart.Diseasecategorical_data <- complete(imputed_categorical)
# Reemplazar las columnas imputadas en el conjunto de datos original
data_antes[, categorical_columns] <- categorical_data
summary(data_antes)## Id NivelSocieconomico Age Sex Admission_Date
## Min. : 1.00 Min. : 0.00 Min. :34.00 Femenino :104 Min. :2010-01-24 00:00:00.00
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 Masculino:216 1st Qu.:2012-05-11 12:00:00.00
## Median :160.50 Median :156.50 Median :56.00 Median :2015-07-26 00:00:00.00
## Mean :160.50 Mean :157.85 Mean :54.99 Mean :2015-07-09 12:07:07.72
## 3rd Qu.:240.25 3rd Qu.:237.25 3rd Qu.:62.00 3rd Qu.:2018-08-21 00:00:00.00
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :2020-12-16 00:00:00.00
## NA's :17
## Chest pain type BP Cholesterol FBS over 120 EKG results Max HR Exercise angina
## 1: 22 Min. : 94.0 Min. :126.0 No:273 0:156 Min. : 71.0 No:213
## 2: 50 1st Qu.:120.0 1st Qu.:212.8 Si: 47 1: 4 1st Qu.:135.5 Si:107
## 3: 98 Median :130.0 Median :244.5 2:160 Median :152.0
## 4:150 Mean :131.3 Mean :248.8 Mean :149.0
## 3rd Qu.:140.0 3rd Qu.:281.2 3rd Qu.:164.2
## Max. :192.0 Max. :417.0 Max. :202.0
##
## ST depression Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B
## Min. :0.0 1:151 Min. :0.0000 3:181 Min. : 1.636 Min. : 1.565
## 1st Qu.:0.0 2:148 1st Qu.:0.0000 6: 18 1st Qu.: 7.508 1st Qu.: 5.494
## Median :0.8 3: 21 Median :0.0000 7:121 Median : 9.660 Median : 6.918
## Mean :1.1 Mean :0.7031 Mean : 11.054 Mean : 7.931
## 3rd Qu.:1.8 3rd Qu.:1.0000 3rd Qu.: 11.751 3rd Qu.: 8.247
## Max. :6.2 Max. :3.0000 Max. :435.045 Max. :350.024
##
## Heart Disease
## Absence :173
## Presence:147
##
##
##
##
## Debido a que las fechas no son consideradas variables categóricas ni numéricas, esta debe ser clasificada como Date, se usa un modelo preliminar con variables posiblemente relacionadas con las fechas de admisión y se predicen los datos faltantes.
data_antes$Admission_Date <- as.Date(data_antes$Admission_Date)
class(data_antes$Admission_Date)## [1] "Date"# Ajustar un modelo de regresión para predecir Admission_Date
model <- lm(as.numeric(Admission_Date) ~ . , data = data_antes, na.action = na.exclude)
# Predecir valores faltantes
predicted_dates <- predict(model, newdata = data_antes[is.na(data_antes$Admission_Date), ])
data_antes$Admission_Date[is.na(data_antes$Admission_Date)] <- as.Date(predicted_dates, origin = "1970-01-01")PASO IV: Analisis de colinealidad de los datos.
Para analizar la colinealidad de los datos se analizó la paquetería Tableone que muestra la colinealidad entra las variables y nos muestra el p-value entre ellas, se seleccionaran las más representativas. La variable de estratificación será “Heart Desease” debido a que esa es la que se quiere analizar
# Obtener todos los nombres de las columnas de data_antes, excepto la variable de estratificación
variables <- setdiff(names(data_antes), "Heart Disease")
# Crear la tabla estratificada por "Heart Disease"
table_one <- CreateTableOne(vars = variables,
strata = "Heart Disease",
data = data_antes,
test = TRUE)## Warning in CreateTableOne(vars = variables, strata = "Heart Disease", data = data_antes, : Dropping variable(s) Admission_Date due to unsupported class.# Mostrar la tabla
print(table_one)## Stratified by Heart Disease
## Absence Presence p test
## n 173 147
## Id (mean (SD)) 151.09 (89.98) 171.57 (94.53) 0.048
## NivelSocieconomico (mean (SD)) 157.92 (92.22) 157.77 (91.51) 0.988
## Age (mean (SD)) 53.77 (9.93) 56.43 (8.42) 0.011
## Sex = Masculino (%) 92 (53.2) 124 (84.4) <0.001
## Chest pain type (%) <0.001
## 1 17 ( 9.8) 5 ( 3.4)
## 2 43 (24.9) 7 ( 4.8)
## 3 70 (40.5) 28 (19.0)
## 4 43 (24.9) 107 (72.8)
## BP (mean (SD)) 129.20 (16.50) 133.80 (17.54) 0.016
## Cholesterol (mean (SD)) 242.81 (46.45) 255.84 (50.13) 0.016
## FBS over 120 = Si (%) 24 (13.9) 23 (15.6) 0.773
## EKG results (%) 0.004
## 0 99 (57.2) 57 (38.8)
## 1 2 ( 1.2) 2 ( 1.4)
## 2 72 (41.6) 88 (59.9)
## Max HR (mean (SD)) 156.84 (19.08) 139.88 (22.51) <0.001
## Exercise angina = Si (%) 28 (16.2) 79 (53.7) <0.001
## ST depression (mean (SD)) 0.67 (0.81) 1.60 (1.30) <0.001
## Slope of ST (%) <0.001
## 1 113 (65.3) 38 (25.9)
## 2 52 (30.1) 96 (65.3)
## 3 8 ( 4.6) 13 ( 8.8)
## Number of vessels fluro (mean (SD)) 0.29 (0.65) 1.18 (1.04) <0.001
## Thallium (%) <0.001
## 3 140 (80.9) 41 (27.9)
## 6 7 ( 4.0) 11 ( 7.5)
## 7 26 (15.0) 95 (64.6)
## BiomarcadorBasal_A (mean (SD)) 9.93 (3.08) 12.37 (35.27) 0.366
## BiomarcadorBasal_B (mean (SD)) 8.89 (26.16) 6.80 (2.03) 0.333# Versión más detallada con niveles y sin comillas
print(table_one, showAllLevels = TRUE, quote = FALSE, nospaces = TRUE)## Stratified by Heart Disease
## level Absence Presence p test
## n 173 147
## Id (mean (SD)) 151.09 (89.98) 171.57 (94.53) 0.048
## NivelSocieconomico (mean (SD)) 157.92 (92.22) 157.77 (91.51) 0.988
## Age (mean (SD)) 53.77 (9.93) 56.43 (8.42) 0.011
## Sex (%) Femenino 81 (46.8) 23 (15.6) <0.001
## Masculino 92 (53.2) 124 (84.4)
## Chest pain type (%) 1 17 ( 9.8) 5 ( 3.4) <0.001
## 2 43 (24.9) 7 ( 4.8)
## 3 70 (40.5) 28 (19.0)
## 4 43 (24.9) 107 (72.8)
## BP (mean (SD)) 129.20 (16.50) 133.80 (17.54) 0.016
## Cholesterol (mean (SD)) 242.81 (46.45) 255.84 (50.13) 0.016
## FBS over 120 (%) No 149 (86.1) 124 (84.4) 0.773
## Si 24 (13.9) 23 (15.6)
## EKG results (%) 0 99 (57.2) 57 (38.8) 0.004
## 1 2 ( 1.2) 2 ( 1.4)
## 2 72 (41.6) 88 (59.9)
## Max HR (mean (SD)) 156.84 (19.08) 139.88 (22.51) <0.001
## Exercise angina (%) No 145 (83.8) 68 (46.3) <0.001
## Si 28 (16.2) 79 (53.7)
## ST depression (mean (SD)) 0.67 (0.81) 1.60 (1.30) <0.001
## Slope of ST (%) 1 113 (65.3) 38 (25.9) <0.001
## 2 52 (30.1) 96 (65.3)
## 3 8 ( 4.6) 13 ( 8.8)
## Number of vessels fluro (mean (SD)) 0.29 (0.65) 1.18 (1.04) <0.001
## Thallium (%) 3 140 (80.9) 41 (27.9) <0.001
## 6 7 ( 4.0) 11 ( 7.5)
## 7 26 (15.0) 95 (64.6)
## BiomarcadorBasal_A (mean (SD)) 9.93 (3.08) 12.37 (35.27) 0.366
## BiomarcadorBasal_B (mean (SD)) 8.89 (26.16) 6.80 (2.03) 0.333Se hace un análisis de Stepwise las variables significativas identificadas en el Tableone y se analiza con “forwart”, “both” y “backward”, al final seleccionamos el modelo con AIC menor y tomamos las variables usadas para entrenar nuestro modelo final por regresión logística.
repre = c( "Age", "Sex", "`Chest pain type`",
"BP", "Cholesterol", "`EKG results`",
"`Max HR`", "`Exercise angina`", "`ST depression`",
"`Slope of ST`", "`Number of vessels fluro`", "Thallium" )
# Crear la fórmula para las variables
formula <- as.formula(paste("`Heart Disease` ~ ", paste(repre, collapse = " + ")))
# Ejecutar el modelo paso a paso con selección hacia adelante
stepwise_model_F <- step(
glm(formula, data = data_antes, family = binomial),
direction = "forward",
trace = FALSE
)
stepwise_model_B <- step(
glm(formula, data = data_antes, family = binomial),
direction = "backward",
trace = FALSE
)
stepwise_model_BO <- step(
glm(formula, data = data_antes, family = binomial),
direction = "both",
trace = FALSE
)
# Obtener los AICs de los modelos
aic_values <- c(
Forward = AIC(stepwise_model_F),
Backward = AIC(stepwise_model_B),
Both = AIC(stepwise_model_BO)
)
# Encontrar el modelo con el menor AIC
mejor_direccion <- names(which.min(aic_values))
mejor_aic <- min(aic_values)
# Mostrar resultados
cat("El modelo con el menor AIC es:", mejor_direccion, "\n")## El modelo con el menor AIC es: Backwardcat("Valor del AIC:", mejor_aic, "\n")## Valor del AIC: 234.789# Resumen del mejor modelo
if (mejor_direccion == "Forward") {
resumen_mejor_modelo <- summary(stepwise_model_F)
} else if (mejor_direccion == "Backward") {
resumen_mejor_modelo <- summary(stepwise_model_B)
} else {
resumen_mejor_modelo <- summary(stepwise_model_BO)
}
print(resumen_mejor_modelo)##
## Call:
## glm(formula = `Heart Disease` ~ Sex + `Chest pain type` + BP +
## `Exercise angina` + `Slope of ST` + `Number of vessels fluro` +
## Thallium, family = binomial, data = data_antes)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -9.31643 1.82564 -5.103 3.34e-07 ***
## SexMasculino 2.17152 0.51576 4.210 2.55e-05 ***
## `Chest pain type`2 1.10403 0.79650 1.386 0.16571
## `Chest pain type`3 0.86497 0.66121 1.308 0.19082
## `Chest pain type`4 2.83473 0.68191 4.157 3.22e-05 ***
## BP 0.02625 0.01012 2.593 0.00950 **
## `Exercise angina`Si 0.84285 0.40921 2.060 0.03943 *
## `Slope of ST`2 1.65041 0.40140 4.112 3.93e-05 ***
## `Slope of ST`3 1.70443 0.74246 2.296 0.02170 *
## `Number of vessels fluro` 1.17369 0.22722 5.165 2.40e-07 ***
## Thallium6 0.23348 0.68646 0.340 0.73376
## Thallium7 1.13477 0.39954 2.840 0.00451 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 441.50 on 319 degrees of freedom
## Residual deviance: 210.79 on 308 degrees of freedom
## AIC: 234.79
##
## Number of Fisher Scoring iterations: 6Parte VI: Prediccion de Heard desease seguun el modelo resultante.
set.seed(123) # Para reproducibilidad
sample_index <- sample(1:nrow(data_antes), size = 0.7 * nrow(data_antes))
train_data <- data_antes[sample_index, ]
test_data <- data_antes[-sample_index, ]
# Asegurarte de que la variable dependiente sea categórica
train_data$Heart_Disease <- as.factor(train_data$`Heart Disease`)
test_data$Heart_Disease <- as.factor(test_data$`Heart Disease`)
# Ajustar el modelo de regresión logística
logistic_model <- glm(`Heart Disease` ~ Sex + `Chest pain type` + BP +
`Exercise angina` + `Slope of ST` + `Number of vessels fluro` +
Thallium,
data = train_data, family = binomial)
# Resumen del modelo
summary(logistic_model)##
## Call:
## glm(formula = `Heart Disease` ~ Sex + `Chest pain type` + BP +
## `Exercise angina` + `Slope of ST` + `Number of vessels fluro` +
## Thallium, family = binomial, data = train_data)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -11.46201 2.60839 -4.394 1.11e-05 ***
## SexMasculino 2.88708 0.78581 3.674 0.000239 ***
## `Chest pain type`2 1.31460 1.15773 1.135 0.256166
## `Chest pain type`3 1.14683 0.89712 1.278 0.201127
## `Chest pain type`4 4.19790 1.01294 4.144 3.41e-05 ***
## BP 0.02819 0.01325 2.128 0.033343 *
## `Exercise angina`Si 0.87941 0.61103 1.439 0.150090
## `Slope of ST`2 2.37502 0.60596 3.919 8.88e-05 ***
## `Slope of ST`3 1.79679 1.02742 1.749 0.080319 .
## `Number of vessels fluro` 1.54646 0.35708 4.331 1.49e-05 ***
## Thallium6 0.53503 1.06388 0.503 0.615029
## Thallium7 1.39992 0.56786 2.465 0.013691 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 310.37 on 223 degrees of freedom
## Residual deviance: 110.40 on 212 degrees of freedom
## AIC: 134.4
##
## Number of Fisher Scoring iterations: 7# Predicciones en el conjunto de prueba
predicciones_prob <- predict(logistic_model, newdata = test_data, type = "response")
# Convertir probabilidades a clases (umbral de 0.5)
predicciones_clases <- ifelse(predicciones_prob >= 0.5, 1, 0)
# Matriz de confusión
matriz_confusion <- table(Predicho = predicciones_clases, Real = test_data$Heart_Disease)
# Extraer los valores de la matriz de confusión
TN <- matriz_confusion[1,1]
FP <- matriz_confusion[1,2]
FN <- matriz_confusion[2,1]
TP <- matriz_confusion[2,2]
# Sensibilidad (Recall o TPR)
sensitivity <- TP / (TP + FN)
# Especificidad (TNR)
specificity <- TN / (TN + FP)
# PPV (Valor Predictivo Positivo)
ppv <- TP / (TP + FP)
# NPV (Valor Predictivo Negativo)
npv <- TN / (TN + FN)
# Mostrar los resultados
cat("Sensibilidad: ", sensitivity, "\n")## Sensibilidad: 0.7435897cat("Especificidad: ", specificity, "\n")## Especificidad: 0.8421053cat("PPV: ", ppv, "\n")## PPV: 0.7631579cat("NPV: ", npv, "\n")## NPV: 0.8275862# Calcular accuracy
accuracy <- sum(diag(matriz_confusion)) / sum(matriz_confusion)
cat("Accuracy del modelo:", accuracy, "\n")## Accuracy del modelo: 0.8020833# Gráfico ROC y AUC
library(pROC)
roc_curve <- roc(test_data$Heart_Disease, predicciones_prob)## Setting levels: control = Absence, case = Presence## Setting direction: controls < casesplot(roc_curve, main = "Curva ROC", col = "blue")
auc <- auc(roc_curve)
cat("Área bajo la curva (AUC):", auc, "\n")## Área bajo la curva (AUC): 0.8477768matriz_confusion <- table(Predicho = predicciones_clases, Real = test_data$Heart_Disease)
# Convertir la matriz de confusión en un data frame para ggplot2
conf_df <- as.data.frame(as.table(matriz_confusion))
# Graficar la matriz de confusión
ggplot(conf_df, aes(x = Predicho, y = Real, fill = Freq)) +
geom_tile(color = "white") +
scale_fill_gradient(low = "lightblue", high = "blue") +
geom_text(aes(label = Freq), vjust = 1, color = "white", size = 5) +
labs(
title = "Matriz de Confusión - Modelo Logístico",
x = "Clase Predicha",
y = "Clase Real"
) +
theme_minimal() +
theme(plot)
metrics_summary <- data.frame(
Metric = c("Sensibilidad", "Especificidad", "PPV", "NPV", "Accuracy", "AUC"),
Value = c(sensitivity, specificity, ppv, npv, accuracy, auc)
)
# Mostrar la tabla resumen
print(metrics_summary)PARTE 2:Analizar, entre los pacientes que han sufrido un IAM y pronóstico a corto y medio plazo, asi como en la respuesta a tratamiento.
Parte I: Seleccion de datos.
Seleccionamos las otra Parte de la base de datos guardadno las variables d einteres en un vector
selected_columns <- c("BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Prognosis_Scale_24h", "Killip_Class_24H", "Recovery_Discharge",
"Treatment_Response", "Treatment_Type", "3M_Prognosis",
"Death_Date", "Cholesterol_Level", "Glucose_Level",
"Neutrophils_Count", "Platelets_Count")
# Select only the desired columns
data_despues <- data[, selected_columns]
# Eliminar la última fila de data_imputed
data_despues <- data_despues[-nrow(data_despues), ]
# Verificar las dimensiones después de eliminar la fila
summary(data_despues)## BiomarcadorAgudo_C Symptom_Scale_Admission Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge
## Min. : 23.92 Min. : 0.00 Min. : 0.00 Length:320 Length:320
## 1st Qu.: 46.37 1st Qu.: 8.00 1st Qu.:16.00 Class :character Class :character
## Median : 60.97 Median :19.00 Median :26.00 Mode :character Mode :character
## Mean : 67.53 Mean :20.53 Mean :26.31
## 3rd Qu.: 79.83 3rd Qu.:34.00 3rd Qu.:40.00
## Max. :218.84 Max. :48.00 Max. :50.00
## NA's :163 NA's :183 NA's :191
## Treatment_Response Treatment_Type 3M_Prognosis Death_Date Cholesterol_Level
## Length:320 Length:320 Length:320 Min. :2010-11-23 00:00:00.00 Min. : 69.01
## Class :character Class :character Class :character 1st Qu.:2015-04-27 00:00:00.00 1st Qu.:170.50
## Mode :character Mode :character Mode :character Median :2019-06-24 00:00:00.00 Median :205.26
## Mean :2018-01-19 01:48:35.41 Mean :204.40
## 3rd Qu.:2019-12-30 00:00:00.00 3rd Qu.:234.01
## Max. :2023-11-27 00:00:00.00 Max. :392.64
## NA's :15 NA's :184
## Glucose_Level Neutrophils_Count Platelets_Count
## Min. : 65.47 Min. :1.890 Min. : 45.62
## 1st Qu.: 87.14 1st Qu.:4.199 1st Qu.:209.48
## Median : 99.34 Median :5.033 Median :247.68
## Mean : 99.02 Mean :5.118 Mean :251.91
## 3rd Qu.:108.45 3rd Qu.:6.015 3rd Qu.:292.82
## Max. :146.18 Max. :8.949 Max. :404.78
## NA's :184 NA's :184 NA's :184*** Paso III: Procesamiento de los datos he imputacion.****
Para tener mejores resultados en la imputación nos aseguramos que las variables categóricas estén clasificadas como un factor.
columns_to_factor <- c("Killip_Class_24H", "Recovery_Discharge", "Treatment_Response", "Treatment_Type", "3M_Prognosis")
existing_columns <- columns_to_factor[columns_to_factor %in% names(data_despues)]
data_despues[existing_columns] <- lapply(data_despues[existing_columns], as.factor)
data_despues$Death_Date <- as.Date(data_despues$Death_Date)
class(data_despues$Death_Date)## [1] "Date"# Ajustar un modelo de regresión para predecir Admission_Date
model <- lm(as.numeric(Death_Date) ~ Treatment_Response + Treatment_Type, data = data_despues, na.action = na.exclude)
# Predecir valores faltantes
predicted_dates <- predict(model, newdata = data_despues[is.na(data_despues$Death_Date), ])
data_despues$Death_Date[is.na(data_despues$Death_Date)] <- as.Date(predicted_dates, origin = "1970-01-01")
lapply(data_despues, class)## $BiomarcadorAgudo_C
## [1] "numeric"
##
## $Symptom_Scale_Admission
## [1] "numeric"
##
## $Prognosis_Scale_24h
## [1] "numeric"
##
## $Killip_Class_24H
## [1] "factor"
##
## $Recovery_Discharge
## [1] "factor"
##
## $Treatment_Response
## [1] "factor"
##
## $Treatment_Type
## [1] "factor"
##
## $`3M_Prognosis`
## [1] "factor"
##
## $Death_Date
## [1] "Date"
##
## $Cholesterol_Level
## [1] "numeric"
##
## $Glucose_Level
## [1] "numeric"
##
## $Neutrophils_Count
## [1] "numeric"
##
## $Platelets_Count
## [1] "numeric"En este caso se imputaros los datos por media y moda, las fechas seran iimputadas con la paqueteria mice
# Procesamiento de datos.
# Función para calcular la moda
calcular_moda <- function(x) {
ux <- unique(x[!is.na(x)]) # Valores únicos excluyendo NA
ux[which.max(tabulate(match(x, ux)))]
}
# Imputación por media y moda
imputar_datos <- function(data_depues) {
for (col in colnames(data_despues)) {
if (is.numeric(data_despues[[col]])) {
# Imputar con la media para variables numéricas
media <- mean(data_despues[[col]], na.rm = TRUE)
data_despues[[col]][is.na(data_despues[[col]])] <- media
} else if (is.factor(data_despues[[col]]) || is.character(data_despues[[col]])) {
# Imputar con la moda para variables categóricas
moda <- calcular_moda(data_despues[[col]])
data_despues[[col]][is.na(data_despues[[col]])] <- moda
}
}
return(data_despues)
}
# Aplicar imputación a los datos
datos_imputados <- imputar_datos(data_despues)
# Verificar la imputación
summary(datos_imputados)## BiomarcadorAgudo_C Symptom_Scale_Admission Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge
## Min. : 23.92 Min. : 0.00 Min. : 0.00 I :259 Good : 39
## 1st Qu.: 61.21 1st Qu.:20.53 1st Qu.:26.31 II : 25 Moderate: 29
## Median : 67.53 Median :20.53 Median :26.31 III: 26 Poor :252
## Mean : 67.53 Mean :20.53 Mean :26.31 IV : 10
## 3rd Qu.: 67.53 3rd Qu.:20.53 3rd Qu.:26.31
## Max. :218.84 Max. :48.00 Max. :50.00
##
## Treatment_Response Treatment_Type 3M_Prognosis Death_Date Cholesterol_Level
## Good : 29 Intervention + Medication: 46 Mild :259 Min. :2010-11-23 Min. : 69.01
## Poor :242 Medication Only :268 Moderate: 26 1st Qu.:2015-05-28 1st Qu.:204.40
## Regular: 49 No Treatment : 6 Severe : 35 Median :2019-05-14 Median :204.40
## Mean :2018-01-20 Mean :204.40
## 3rd Qu.:2019-12-30 3rd Qu.:204.40
## Max. :2023-11-27 Max. :392.64
## NA's :8
## Glucose_Level Neutrophils_Count Platelets_Count
## Min. : 65.47 Min. :1.890 Min. : 45.62
## 1st Qu.: 99.02 1st Qu.:5.118 1st Qu.:251.91
## Median : 99.02 Median :5.118 Median :251.91
## Mean : 99.02 Mean :5.118 Mean :251.91
## 3rd Qu.: 99.02 3rd Qu.:5.118 3rd Qu.:251.91
## Max. :146.18 Max. :8.949 Max. :404.78
## Imputación de fechas (en este caso Death_Date)
data_despues\(Death_Date <- as.numeric(as.Date(data_despues\)Death_Date))
Usar mice para la imputación
library(mice) meth <- rep(“rf”, ncol(data_despues)) # Usa Random Forest para la imputación imp <- mice(data_despues, method = meth)
Recuperar los datos imputados
data_imputed <- complete(imp)
Unimos las bases de datos en la arte 1 y 2 para tener encuentra todas las variables y filtramos para tener solo a los pacientes con presencia de enfermedades cardiacas
# Unir ambas bases de datos
data_total <- cbind(data_antes, data_imputed)
summary(data_total)## Id NivelSocieconomico Age Sex Admission_Date Chest pain type
## Min. : 1.00 Min. : 0.00 Min. :34.00 Femenino :104 Min. :2010-01-24 1: 22
## 1st Qu.: 80.75 1st Qu.: 78.75 1st Qu.:48.00 Masculino:216 1st Qu.:2012-06-27 2: 50
## Median :160.50 Median :156.50 Median :56.00 Median :2016-01-20 3: 98
## Mean :160.50 Mean :157.85 Mean :54.99 Mean :2015-11-06 4:150
## 3rd Qu.:240.25 3rd Qu.:237.25 3rd Qu.:62.00 3rd Qu.:2019-01-13
## Max. :320.00 Max. :319.00 Max. :79.00 Max. :2022-05-07
## BP Cholesterol FBS over 120 EKG results Max HR Exercise angina ST depression
## Min. : 94.0 Min. :126.0 No:273 0:156 Min. : 71.0 No:213 Min. :0.0
## 1st Qu.:120.0 1st Qu.:212.8 Si: 47 1: 4 1st Qu.:135.5 Si:107 1st Qu.:0.0
## Median :130.0 Median :244.5 2:160 Median :152.0 Median :0.8
## Mean :131.3 Mean :248.8 Mean :149.0 Mean :1.1
## 3rd Qu.:140.0 3rd Qu.:281.2 3rd Qu.:164.2 3rd Qu.:1.8
## Max. :192.0 Max. :417.0 Max. :202.0 Max. :6.2
## Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B Heart Disease
## 1:151 Min. :0.0000 3:181 Min. : 1.636 Min. : 1.565 Absence :173
## 2:148 1st Qu.:0.0000 6: 18 1st Qu.: 7.508 1st Qu.: 5.494 Presence:147
## 3: 21 Median :0.0000 7:121 Median : 9.660 Median : 6.918
## Mean :0.7031 Mean : 11.054 Mean : 7.931
## 3rd Qu.:1.0000 3rd Qu.: 11.751 3rd Qu.: 8.247
## Max. :3.0000 Max. :435.045 Max. :350.024
## BiomarcadorAgudo_C Symptom_Scale_Admission Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge
## Min. : 23.92 Min. : 0.00 Min. : 0.00 I :227 Good : 97
## 1st Qu.: 46.88 1st Qu.: 8.00 1st Qu.:16.00 II : 38 Moderate: 53
## Median : 59.92 Median :21.00 Median :26.00 III: 44 Poor :170
## Mean : 66.29 Mean :20.85 Mean :26.40 IV : 11
## 3rd Qu.: 78.39 3rd Qu.:34.00 3rd Qu.:39.25
## Max. :218.84 Max. :48.00 Max. :50.00
## Treatment_Response Treatment_Type 3M_Prognosis Death_Date Cholesterol_Level
## Good : 58 Intervention + Medication:100 Mild :227 Min. :14936 Min. : 69.01
## Poor :138 Medication Only :208 Moderate: 37 1st Qu.:16628 1st Qu.:167.04
## Regular:124 No Treatment : 12 Severe : 56 Median :18058 Median :204.34
## Mean :17563 Mean :202.82
## 3rd Qu.:18260 3rd Qu.:233.98
## Max. :19688 Max. :392.64
## Glucose_Level Neutrophils_Count Platelets_Count
## Min. : 65.47 Min. :1.890 Min. : 45.62
## 1st Qu.: 87.62 1st Qu.:4.213 1st Qu.:211.66
## Median : 99.17 Median :5.057 Median :247.36
## Mean : 98.59 Mean :5.164 Mean :254.58
## 3rd Qu.:107.38 3rd Qu.:6.092 3rd Qu.:292.54
## Max. :146.18 Max. :8.949 Max. :404.78data_total <- data_total %>% filter(`Heart Disease` == "Presence")Nos aseguramos que las variables categoricas esten como factores y si es posible unificamos variables teniendo encuenta su modleo teorico y la estadistica de sus datos
# Función para convertir números romanos a arábigos (definida anteriormente)
data_total$Killip_Class_24H <- as.numeric(data_total$Killip_Class_24H)
class(data_total$Killip_Class_24H)## [1] "numeric"# Unificar los niveles "leve" y "mild"
data_total$`3M_Prognosis`## [1] Moderate Mild Mild Severe Severe Moderate Mild Mild Severe Mild Mild
## [12] Mild Mild Moderate Mild Mild Mild Severe Mild Moderate Severe Moderate
## [23] Moderate Moderate Moderate Mild Moderate Mild Moderate Moderate Mild Severe Severe
## [34] Moderate Moderate Mild Mild Mild Severe Mild Moderate Severe Mild Mild
## [45] Moderate Mild Mild Mild Severe Mild Mild Mild Mild Mild Mild
## [56] Mild Mild Mild Moderate Severe Severe Mild Mild Moderate Mild Mild
## [67] Mild Mild Severe Severe Mild Severe Mild Severe Mild Mild Mild
## [78] Mild Mild Mild Severe Mild Severe Mild Moderate Mild Mild Mild
## [89] Moderate Moderate Mild Mild Moderate Mild Mild Mild Severe Mild Mild
## [100] Mild Mild Mild Mild Severe Moderate Moderate Severe Severe Mild Severe
## [111] Severe Severe Severe Mild Mild Mild Severe Mild Moderate Mild Mild
## [122] Mild Moderate Mild Severe Moderate Mild Mild Severe Severe Severe Severe
## [133] Mild Mild Severe Mild Mild Severe Severe Mild Mild Severe Severe
## [144] Mild Severe Mild Mild
## Levels: Mild Moderate Severedata_total$`3M_Prognosis` <- factor(data_total$`3M_Prognosis`,
levels = c("Mild", "Leve", "Severe"),
labels = c("Leve_mild", "Leve_mild", "Severe"))
levels(data_total$data_total$`3M_Prognosis`)## NULL# Contar las observaciones por nivel
table(data_total$data_total$`3M_Prognosis`)## < table of extent 0 >data_total$data_total$`3M_Prognosis` <- trimws(data_total$`3M_Prognosis`)data_total <- data_total %>%
mutate(`Chest pain type` = case_when(
`Chest pain type` %in% c(1, 2) ~ "leve",
`Chest pain type` %in% c(3, 4) ~ "agudo"
))
# Instalar el paquete forcats si no lo tienes
# install.packages("forcats")
library(forcats)
# Recodificar la variable 'recobre_discharch'
data_total$Recovery_Discharge <- fct_recode(data_total$Recovery_Discharge,
"good_moderate" = "Good",
"good_moderate" = "Moderate")
# Ver los niveles de la variable después de la recodificación
levels(data_total$Recovery_Discharge)## [1] "good_moderate" "Poor"factores <- c("Chest pain type","Recovery_Discharge")
existing_columns <- factores[factores %in% names(data_total)]
data_total[existing_columns] <- lapply(data_total[existing_columns], as.factor)
summary(data_total)## Id NivelSocieconomico Age Sex Admission_Date Chest pain type
## Min. : 1.0 Min. : 0.0 Min. :35.00 Femenino : 23 Min. :2010-01-24 agudo:135
## 1st Qu.: 85.0 1st Qu.: 85.5 1st Qu.:51.00 Masculino:124 1st Qu.:2012-10-14 leve : 12
## Median :177.0 Median :148.0 Median :58.00 Median :2016-08-24
## Mean :171.6 Mean :157.8 Mean :56.43 Mean :2016-04-06
## 3rd Qu.:258.0 3rd Qu.:238.5 3rd Qu.:62.00 3rd Qu.:2019-06-02
## Max. :319.0 Max. :319.0 Max. :79.00 Max. :2021-11-10
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## BP Cholesterol FBS over 120 EKG results Max HR Exercise angina ST depression
## Min. :100.0 Min. :141.0 No:124 0:57 Min. : 71.0 No:68 Min. :0.000
## 1st Qu.:120.0 1st Qu.:226.5 Si: 23 1: 2 1st Qu.:125.0 Si:79 1st Qu.:0.600
## Median :132.0 Median :255.0 2:88 Median :142.0 Median :1.400
## Mean :133.8 Mean :255.8 Mean :139.9 Mean :1.602
## 3rd Qu.:141.0 3rd Qu.:288.0 3rd Qu.:156.5 3rd Qu.:2.400
## Max. :192.0 Max. :409.0 Max. :195.0 Max. :6.200
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B Heart Disease
## 1:38 Min. :0.000 3:41 Min. : 1.636 Min. : 1.835 Absence : 0
## 2:96 1st Qu.:0.000 6:11 1st Qu.: 6.948 1st Qu.: 5.499 Presence:147
## 3:13 Median :1.000 7:95 Median : 9.425 Median : 6.836
## Mean :1.184 Mean : 12.371 Mean : 6.797
## 3rd Qu.:2.000 3rd Qu.: 11.624 3rd Qu.: 8.352
## Max. :3.000 Max. :435.045 Max. :11.662
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## BiomarcadorAgudo_C Symptom_Scale_Admission Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge
## Min. : 23.92 Min. : 0.00 Min. : 0.00 Min. :1.000 good_moderate:72
## 1st Qu.: 46.16 1st Qu.: 8.00 1st Qu.:14.50 1st Qu.:1.000 Poor :75
## Median : 62.65 Median :19.00 Median :26.00 Median :1.000
## Mean : 66.73 Mean :20.37 Mean :25.94 Mean :1.748
## 3rd Qu.: 79.12 3rd Qu.:34.00 3rd Qu.:40.00 3rd Qu.:2.500
## Max. :218.84 Max. :48.00 Max. :50.00 Max. :4.000
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Treatment_Response Treatment_Type 3M_Prognosis Death_Date Cholesterol_Level
## Good :31 Intervention + Medication:47 Leve_mild:83 Min. :15049 Min. : 69.01
## Poor :62 Medication Only :93 Severe :38 1st Qu.:16773 1st Qu.:170.48
## Regular:54 No Treatment : 7 NA's :26 Median :18173 Median :205.55
## Mean :17673 Mean :204.27
## 3rd Qu.:18260 3rd Qu.:233.54
## Max. :19688 Max. :392.64
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Glucose_Level Neutrophils_Count Platelets_Count data_total.Length data_total.Class data_total.Mode
## Min. : 65.47 Min. :1.890 Min. : 45.62 147 -none- character
## 1st Qu.: 87.21 1st Qu.:4.208 1st Qu.:209.59 147 -none- character
## Median : 98.28 Median :5.042 Median :247.77 147 -none- character
## Mean : 98.50 Mean :5.149 Mean :253.25 147 -none- character
## 3rd Qu.:107.94 3rd Qu.:6.023 3rd Qu.:293.10 147 -none- character
## Max. :146.18 Max. :8.949 Max. :404.78 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## [ reached getOption("max.print") -- omitted 117 rows ]Analisamos la correlacion de los datos
library(corrplot)
cor_data <- data_total %>%
select_if(is.numeric) %>%
cor(use = "complete.obs")
corrplot(cor_data, method = "circle")
PASO IV: Analisis de colinealidad de los datos.
Para analizar la colinealidad de los datos se analizó la paquetería Tableone que muestra la colinealidad entra las variables y nos muestra el p-value entre ellas, se seleccionaran las más representativas. La variable de estratificación será “3M_Prognosis” debido a que esa es la que se quiere analizar.
# Variables seleccionadas para la tabla
selected_vars <- c(
"Age", "Sex", "`Chest pain type`", "BP", "Cholesterol",
"`EKG results`", "`Max HR`", "`Exercise angina`",
"`ST depression`", "`Slope of ST`", "`Number of vessels fluro`",
"Thallium", "BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Killip_Class_24H", "Recovery_Discharge", "Treatment_Response",
"Treatment_Type", "Cholesterol_Level", "Glucose_Level",
"Neutrophils_Count", "Platelets_Count","BiomarcadorBasal_A","BiomarcadorBasal_B"
)
# Verificar que las variables existen en el conjunto de datos
selected_vars <- intersect(selected_vars, names(data_total))
# Crear la Table One estratificada por 3M_Prognosis
table_one <- CreateTableOne(
vars = selected_vars,
strata = "3M_Prognosis", # Variable de estratificación
data = data_total,
factorVars = selected_vars[sapply(data_total[selected_vars], is.factor)]
)
# Mostrar la tabla en consola
print(table_one, showAllLevels = TRUE)## Stratified by 3M_Prognosis
## level Leve_mild Severe p test
## n 83 38
## Age (mean (SD)) 56.84 (8.48) 56.16 (9.09) 0.687
## Sex (%) Femenino 8 ( 9.6) 9 (23.7) 0.075
## Masculino 75 (90.4) 29 (76.3)
## BP (mean (SD)) 133.52 (16.35) 138.84 (18.75) 0.115
## Cholesterol (mean (SD)) 252.06 (48.89) 263.68 (59.70) 0.260
## Thallium (%) 3 22 (26.5) 7 (18.4) 0.420
## 6 6 ( 7.2) 5 (13.2)
## 7 55 (66.3) 26 (68.4)
## BiomarcadorAgudo_C (mean (SD)) 65.55 (30.36) 79.69 (28.72) 0.017
## Symptom_Scale_Admission (mean (SD)) 22.72 (12.80) 13.32 (13.29) <0.001
## Killip_Class_24H (mean (SD)) 1.00 (0.00) 3.21 (0.58) <0.001
## Recovery_Discharge (%) good_moderate 33 (39.8) 26 (68.4) 0.006
## Poor 50 (60.2) 12 (31.6)
## Treatment_Response (%) Good 13 (15.7) 9 (23.7) 0.528
## Poor 38 (45.8) 17 (44.7)
## Regular 32 (38.6) 12 (31.6)
## Treatment_Type (%) Intervention + Medication 30 (36.1) 6 (15.8) 0.003
## Medication Only 52 (62.7) 27 (71.1)
## No Treatment 1 ( 1.2) 5 (13.2)
## Cholesterol_Level (mean (SD)) 203.52 (49.97) 200.20 (58.56) 0.749
## Glucose_Level (mean (SD)) 98.37 (16.03) 100.40 (13.59) 0.499
## Neutrophils_Count (mean (SD)) 5.08 (1.48) 5.15 (1.14) 0.819
## Platelets_Count (mean (SD)) 248.37 (56.23) 252.59 (61.72) 0.711
## BiomarcadorBasal_A (mean (SD)) 14.50 (46.86) 9.62 (3.05) 0.523
## BiomarcadorBasal_B (mean (SD)) 7.03 (2.05) 6.39 (1.96) 0.108Se hace un análisis de Stepwise las variables significativas identificadas en el Tableone y se analiza con “forwart”, “both” y “backward”, al final seleccionamos el modelo con AIC menor y tomamos las variables usadas para entrenar nuestro modelo final por regresión logística.
# Variables seleccionadas
selected_vars <- c(
"Sex", "`Chest pain type`", "BP",
"`EKG results`", "`Max HR`", "`Exercise angina`",
"`ST depression`", "`Slope of ST`", "`Number of vessels fluro`",
"Thallium", "BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Recovery_Discharge",
"Treatment_Type", "Death_Date",
"Glucose_Level", "Platelets_Count","BiomarcadorBasal_B"
)
# Filtrar variables existentes en el conjunto de datos
repre <- intersect(selected_vars, names(data_total))
# Remover la variable objetivo del conjunto `repre`
repre <- repre[repre != "3M_Prognosis"]
# Crear la fórmula con las variables seleccionadas
formula <- as.formula(paste("`3M_Prognosis` ~", paste(repre, collapse = " + ")))
# Remover NAs del conjunto de datos
data_total <- na.omit(data_total)
# Ejecutar los modelos stepwise
stepwise_model_F <- step(
glm(formula, data = data_total, family = binomial),
direction = "forward",
trace = FALSE
)
stepwise_model_B <- step(
glm(formula, data = data_total, family = binomial),
direction = "backward",
trace = FALSE
)
stepwise_model_BO <- step(
glm(formula, data = data_total, family = binomial),
direction = "both",
trace = FALSE
)
# Obtener los AICs de los modelos
aic_values <- c(
Forward = AIC(stepwise_model_F),
Backward = AIC(stepwise_model_B),
Both = AIC(stepwise_model_BO)
)
# Encontrar el modelo con el menor AIC
mejor_direccion <- names(which.min(aic_values))
mejor_aic <- min(aic_values)
# Mostrar resultados
cat("El modelo con el menor AIC es:", mejor_direccion, "\n")## El modelo con el menor AIC es: Backwardcat("Valor del AIC:", mejor_aic, "\n")## Valor del AIC: 121.4268# Resumen del mejor modelo
if (mejor_direccion == "Forward") {
resumen_mejor_modelo <- summary(stepwise_model_F)
} else if (mejor_direccion == "Backward") {
resumen_mejor_modelo <- summary(stepwise_model_B)
} else {
resumen_mejor_modelo <- summary(stepwise_model_BO)
}
print(resumen_mejor_modelo)##
## Call:
## glm(formula = `3M_Prognosis` ~ Sex + BiomarcadorAgudo_C + Symptom_Scale_Admission +
## Treatment_Type + Death_Date + BiomarcadorBasal_B, family = binomial,
## data = data_total)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.489e+01 6.236e+00 -2.388 0.016933 *
## SexMasculino -1.407e+00 6.935e-01 -2.029 0.042415 *
## BiomarcadorAgudo_C 1.748e-02 7.722e-03 2.264 0.023600 *
## Symptom_Scale_Admission -7.073e-02 1.977e-02 -3.577 0.000348 ***
## Treatment_TypeMedication Only 1.748e+00 6.381e-01 2.740 0.006145 **
## Treatment_TypeNo Treatment 3.398e+00 1.530e+00 2.220 0.026398 *
## Death_Date 8.388e-04 3.367e-04 2.491 0.012735 *
## BiomarcadorBasal_B -1.822e-01 1.245e-01 -1.464 0.143254
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 150.60 on 120 degrees of freedom
## Residual deviance: 105.43 on 113 degrees of freedom
## AIC: 121.43
##
## Number of Fisher Scoring iterations: 5Parte VI: Prediccion de Heard desease seguun el modelo resultante.
set.seed(123) # Para reproducibilidad
sample_index <- sample(1:nrow(data_antes), size = 0.7 * nrow(data_antes))
train_data <- data_total[sample_index, ]
test_data <- data_total[-sample_index, ]
# Asegurarte de que la variable dependiente sea categórica
train_data$`3M_Prognosis`<- as.factor(train_data$`3M_Prognosis`)
test_data$`3M_Prognosis` <- as.factor(test_data$`3M_Prognosis`)
# Ajustar el modelo de regresión logística
logistic_model <- glm(`3M_Prognosis` ~ Sex + BiomarcadorAgudo_C + Symptom_Scale_Admission +
Treatment_Type + Death_Date + BiomarcadorBasal_B,
data = train_data, family = binomial)
# Resumen del modelo
summary(logistic_model)##
## Call:
## glm(formula = `3M_Prognosis` ~ Sex + BiomarcadorAgudo_C + Symptom_Scale_Admission +
## Treatment_Type + Death_Date + BiomarcadorBasal_B, family = binomial,
## data = train_data)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.488e+01 7.248e+00 -2.053 0.040048 *
## SexMasculino -2.042e+00 8.809e-01 -2.318 0.020476 *
## BiomarcadorAgudo_C 2.814e-02 1.209e-02 2.327 0.019942 *
## Symptom_Scale_Admission -1.013e-01 2.721e-02 -3.721 0.000198 ***
## Treatment_TypeMedication Only 1.509e+00 7.819e-01 1.929 0.053680 .
## Treatment_TypeNo Treatment 2.750e+00 1.955e+00 1.406 0.159680
## Death_Date 8.652e-04 3.905e-04 2.216 0.026720 *
## BiomarcadorBasal_B -1.864e-01 1.660e-01 -1.123 0.261285
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 107.771 on 86 degrees of freedom
## Residual deviance: 66.192 on 79 degrees of freedom
## (137 observations deleted due to missingness)
## AIC: 82.192
##
## Number of Fisher Scoring iterations: 5# Predicciones en el conjunto de prueba
predicciones_prob <- predict(logistic_model, newdata = test_data, type = "response")
# Convertir probabilidades a clases (umbral de 0.5)
predicciones_clases <- ifelse(predicciones_prob >= 0.5, 1, 0)
# Matriz de confusión
matriz_confusion <- table(Predicho = predicciones_clases, Real = test_data$`3M_Prognosis`)
print(matriz_confusion)## Real
## Predicho Leve_mild Severe
## 0 17 6
## 1 6 5# Extraer los valores de la matriz de confusión
TN <- matriz_confusion[1,1]
FP <- matriz_confusion[1,2]
FN <- matriz_confusion[2,1]
TP <- matriz_confusion[2,2]
# Sensibilidad (Recall o TPR)
sensitivity <- TP / (TP + FN)
# Especificidad (TNR)
specificity <- TN / (TN + FP)
# PPV (Valor Predictivo Positivo)
ppv <- TP / (TP + FP)
# NPV (Valor Predictivo Negativo)
npv <- TN / (TN + FN)
# Mostrar los resultados
cat("Sensibilidad: ", sensitivity, "\n")## Sensibilidad: 0.4545455cat("Especificidad: ", specificity, "\n")## Especificidad: 0.7391304cat("PPV: ", ppv, "\n")## PPV: 0.4545455cat("NPV: ", npv, "\n")## NPV: 0.7391304# Calcular accuracy
accuracy <- sum(diag(matriz_confusion)) / sum(matriz_confusion)
cat("Accuracy del modelo:", accuracy, "\n")## Accuracy del modelo: 0.6470588# Gráfico ROC y AUC
library(pROC)
roc_curve <- roc(test_data$`3M_Prognosis`, predicciones_prob)## Setting levels: control = Leve_mild, case = Severe## Setting direction: controls < casesplot(roc_curve, main = "Curva ROC", col = "blue")
auc <- auc(roc_curve)
cat("Área bajo la curva (AUC):", auc, "\n")## Área bajo la curva (AUC): 0.7193676library(ggplot2)
# Crear matriz de confusión
matriz_confusion <- table(Predicho = predicciones_clases, Real = test_data$`3M_Prognosis`)
# Verificar matriz de confusión
print(matriz_confusion)## Real
## Predicho Leve_mild Severe
## 0 17 6
## 1 6 5# Convertir la matriz de confusión en un data frame para ggplot2
conf_df <- as.data.frame(as.table(matriz_confusion))
# Graficar la matriz de confusión
ggplot(conf_df, aes(x = Predicho, y = Real, fill = Freq)) +
geom_tile(color = "white") +
scale_fill_gradient(low = "lightpink", high = "red") +
geom_text(aes(label = Freq), vjust = 1, color = "white", size = 5) +
labs(
title = "Matriz de Confusión - Modelo Logístico",
x = "Clase Predicha",
y = "Clase Real"
) +
theme_minimal() +
theme(plot.title = element_text(hjust = 0.5, size = 14))
metrics_summary <- data.frame(
Metric = c("Sensibilidad", "Especificidad", "PPV", "NPV", "Accuracy", "AUC"),
Value = c(sensitivity, specificity, ppv, npv, accuracy, auc)
)
# Mostrar la tabla resumen
print(metrics_summary)PARTE 2,2:Analizar, entre los pacientes que han sufrido un IAM y pronóstico a corto y medio plazo, asi como en la respuesta a tratamiento.
# Variables seleccionadas para la tabla
selected_vars <- c(
"Age", "Sex", "`Chest pain type`", "BP", "Cholesterol",
"`EKG results`", "`Max HR`", "`Exercise angina`",
"`ST depression`", "`Slope of ST`", "`Number of vessels fluro`",
"Thallium", "BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Killip_Class_24H", "Recovery_Discharge", "Treatment_Response",
"Treatment_Type", "Cholesterol_Level", "Glucose_Level",
"Neutrophils_Count", "Platelets_Count","BiomarcadorBasal_A","BiomarcadorBasal_B"
)
# Verificar que las variables existen en el conjunto de datos
selected_vars <- intersect(selected_vars, names(data_total))
# Crear la Table One estratificada por 3M_Prognosis
table_one <- CreateTableOne(
vars = selected_vars,
strata = "3M_Prognosis", # Variable de estratificación
data = data_total,
factorVars = selected_vars[sapply(data_total[selected_vars], is.factor)]
)
# Mostrar la tabla en consola
print(table_one, showAllLevels = TRUE)## Stratified by 3M_Prognosis
## level Leve_mild Severe p test
## n 83 38
## Age (mean (SD)) 56.84 (8.48) 56.16 (9.09) 0.687
## Sex (%) Femenino 8 ( 9.6) 9 (23.7) 0.075
## Masculino 75 (90.4) 29 (76.3)
## BP (mean (SD)) 133.52 (16.35) 138.84 (18.75) 0.115
## Cholesterol (mean (SD)) 252.06 (48.89) 263.68 (59.70) 0.260
## Thallium (%) 3 22 (26.5) 7 (18.4) 0.420
## 6 6 ( 7.2) 5 (13.2)
## 7 55 (66.3) 26 (68.4)
## BiomarcadorAgudo_C (mean (SD)) 65.55 (30.36) 79.69 (28.72) 0.017
## Symptom_Scale_Admission (mean (SD)) 22.72 (12.80) 13.32 (13.29) <0.001
## Killip_Class_24H (mean (SD)) 1.00 (0.00) 3.21 (0.58) <0.001
## Recovery_Discharge (%) good_moderate 33 (39.8) 26 (68.4) 0.006
## Poor 50 (60.2) 12 (31.6)
## Treatment_Response (%) Good 13 (15.7) 9 (23.7) 0.528
## Poor 38 (45.8) 17 (44.7)
## Regular 32 (38.6) 12 (31.6)
## Treatment_Type (%) Intervention + Medication 30 (36.1) 6 (15.8) 0.003
## Medication Only 52 (62.7) 27 (71.1)
## No Treatment 1 ( 1.2) 5 (13.2)
## Cholesterol_Level (mean (SD)) 203.52 (49.97) 200.20 (58.56) 0.749
## Glucose_Level (mean (SD)) 98.37 (16.03) 100.40 (13.59) 0.499
## Neutrophils_Count (mean (SD)) 5.08 (1.48) 5.15 (1.14) 0.819
## Platelets_Count (mean (SD)) 248.37 (56.23) 252.59 (61.72) 0.711
## BiomarcadorBasal_A (mean (SD)) 14.50 (46.86) 9.62 (3.05) 0.523
## BiomarcadorBasal_B (mean (SD)) 7.03 (2.05) 6.39 (1.96) 0.108summary(data_total)## Id NivelSocieconomico Age Sex Admission_Date Chest pain type
## Min. : 4.0 Min. : 0.0 Min. :35.00 Femenino : 17 Min. :2010-03-21 agudo:111
## 1st Qu.:112.0 1st Qu.: 82.0 1st Qu.:53.00 Masculino:104 1st Qu.:2014-01-16 leve : 10
## Median :190.0 Median :147.0 Median :58.00 Median :2017-04-21
## Mean :180.5 Mean :157.7 Mean :56.63 Mean :2016-08-13
## 3rd Qu.:264.0 3rd Qu.:245.0 3rd Qu.:61.00 3rd Qu.:2019-09-04
## Max. :319.0 Max. :319.0 Max. :79.00 Max. :2021-11-10
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## BP Cholesterol FBS over 120 EKG results Max HR Exercise angina ST depression
## Min. :100.0 Min. :149.0 No:103 0:43 Min. : 71.0 No:52 Min. :0.000
## 1st Qu.:122.0 1st Qu.:219.0 Si: 18 1: 2 1st Qu.:125.0 Si:69 1st Qu.:0.600
## Median :134.0 Median :256.0 2:76 Median :140.0 Median :1.500
## Mean :135.2 Mean :255.7 Mean :137.8 Mean :1.645
## 3rd Qu.:145.0 3rd Qu.:288.0 3rd Qu.:154.0 3rd Qu.:2.400
## Max. :192.0 Max. :409.0 Max. :195.0 Max. :6.200
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Slope of ST Number of vessels fluro Thallium BiomarcadorBasal_A BiomarcadorBasal_B Heart Disease
## 1:31 Min. :0.000 3:29 Min. : 1.636 Min. : 1.835 Absence : 0
## 2:79 1st Qu.:0.000 6:11 1st Qu.: 7.041 1st Qu.: 5.550 Presence:121
## 3:11 Median :1.000 7:81 Median : 9.425 Median : 6.815
## Mean :1.165 Mean : 12.972 Mean : 6.828
## 3rd Qu.:2.000 3rd Qu.: 11.705 3rd Qu.: 8.382
## Max. :3.000 Max. :435.045 Max. :11.662
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## BiomarcadorAgudo_C Symptom_Scale_Admission Prognosis_Scale_24h Killip_Class_24H Recovery_Discharge
## Min. : 25.04 Min. : 0.00 Min. : 0.00 Min. :1.000 good_moderate:59
## 1st Qu.: 49.13 1st Qu.: 8.00 1st Qu.:16.00 1st Qu.:1.000 Poor :62
## Median : 65.93 Median :17.00 Median :28.00 Median :1.000
## Mean : 69.99 Mean :19.77 Mean :27.14 Mean :1.694
## 3rd Qu.: 83.73 3rd Qu.:32.00 3rd Qu.:40.00 3rd Qu.:3.000
## Max. :218.84 Max. :48.00 Max. :50.00 Max. :4.000
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Treatment_Response Treatment_Type 3M_Prognosis Death_Date Cholesterol_Level
## Good :22 Intervention + Medication:36 Leve_mild:83 Min. :15049 Min. : 69.01
## Poor :55 Medication Only :79 Severe :38 1st Qu.:17593 1st Qu.:167.33
## Regular:44 No Treatment : 6 Median :18260 Median :203.43
## Mean :17858 Mean :202.48
## 3rd Qu.:18260 3rd Qu.:233.11
## Max. :19688 Max. :392.64
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
##
## Glucose_Level Neutrophils_Count Platelets_Count data_total.Length data_total.Class data_total.Mode
## Min. : 66.83 Min. :1.890 Min. : 76.22 147 -none- character
## 1st Qu.: 87.54 1st Qu.:4.183 1st Qu.:209.59 147 -none- character
## Median : 98.37 Median :5.021 Median :246.54 147 -none- character
## Mean : 99.01 Mean :5.104 Mean :249.70 147 -none- character
## 3rd Qu.:108.15 3rd Qu.:6.022 3rd Qu.:288.83 147 -none- character
## Max. :146.18 Max. :8.949 Max. :389.16 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## 147 -none- character
## [ reached getOption("max.print") -- omitted 91 rows ]# Variables seleccionadas para la tabla
selected_vars <- c(
"Age", "Sex", "`Chest pain type`", "BP", "Cholesterol",
"`EKG results`", "`Max HR`", "`Exercise angina`",
"`ST depression`", "`Slope of ST`", "`Number of vessels fluro`",
"Thallium", "BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Killip_Class_24H", "Recovery_Discharge", "3M_Prognosis",
"Treatment_Type", "Cholesterol_Level", "Glucose_Level",
"Neutrophils_Count", "Platelets_Count","BiomarcadorBasal_A","BiomarcadorBasal_B"
)
# Verificar que las variables existen en el conjunto de datos
selected_vars <- intersect(selected_vars, names(data_total))
# Crear la Table One estratificada por "Treatment_Response"
table_one <- CreateTableOne(
vars = selected_vars,
strata = "Treatment_Response", # Variable de estratificación
data = data_total,
factorVars = selected_vars[sapply(data_total[selected_vars], is.factor)]
)
# Mostrar la tabla en consola
print(table_one, showAllLevels = TRUE)## Stratified by Treatment_Response
## level Good Poor Regular
## n 22 55 44
## Age (mean (SD)) 58.64 (9.60) 56.56 (8.48) 55.70 (8.40)
## Sex (%) Femenino 4 (18.2) 6 (10.9) 7 (15.9)
## Masculino 18 (81.8) 49 (89.1) 37 (84.1)
## BP (mean (SD)) 137.86 (19.31) 133.22 (14.39) 136.32 (19.43)
## Cholesterol (mean (SD)) 247.82 (49.24) 257.87 (50.41) 256.95 (57.38)
## Thallium (%) 3 6 (27.3) 13 (23.6) 10 (22.7)
## 6 3 (13.6) 6 (10.9) 2 ( 4.5)
## 7 13 (59.1) 36 (65.5) 32 (72.7)
## BiomarcadorAgudo_C (mean (SD)) 55.56 (31.43) 79.44 (25.60) 65.39 (32.25)
## Symptom_Scale_Admission (mean (SD)) 22.23 (11.86) 21.35 (14.52) 16.57 (12.96)
## Killip_Class_24H (mean (SD)) 1.91 (1.15) 1.73 (1.13) 1.55 (0.98)
## Recovery_Discharge (%) good_moderate 9 (40.9) 22 (40.0) 28 (63.6)
## Poor 13 (59.1) 33 (60.0) 16 (36.4)
## 3M_Prognosis (%) Leve_mild 13 (59.1) 38 (69.1) 32 (72.7)
## Severe 9 (40.9) 17 (30.9) 12 (27.3)
## Treatment_Type (%) Intervention + Medication 2 ( 9.1) 17 (30.9) 17 (38.6)
## Medication Only 19 (86.4) 35 (63.6) 25 (56.8)
## No Treatment 1 ( 4.5) 3 ( 5.5) 2 ( 4.5)
## Cholesterol_Level (mean (SD)) 198.16 (62.48) 206.09 (50.06) 200.12 (51.32)
## Glucose_Level (mean (SD)) 104.06 (15.45) 96.47 (15.92) 99.65 (13.95)
## Neutrophils_Count (mean (SD)) 4.90 (1.24) 5.12 (1.40) 5.18 (1.43)
## Platelets_Count (mean (SD)) 248.81 (54.55) 247.55 (58.64) 252.82 (59.42)
## BiomarcadorBasal_A (mean (SD)) 8.72 (3.63) 9.12 (3.45) 19.91 (64.11)
## BiomarcadorBasal_B (mean (SD)) 7.25 (2.08) 6.45 (2.06) 7.09 (1.94)
## Stratified by Treatment_Response
## p test
## n
## Age (mean (SD)) 0.433
## Sex (%) 0.642
##
## BP (mean (SD)) 0.491
## Cholesterol (mean (SD)) 0.739
## Thallium (%) 0.686
##
##
## BiomarcadorAgudo_C (mean (SD)) 0.003
## Symptom_Scale_Admission (mean (SD)) 0.144
## Killip_Class_24H (mean (SD)) 0.418
## Recovery_Discharge (%) 0.047
##
## 3M_Prognosis (%) 0.528
##
## Treatment_Type (%) 0.171
##
##
## Cholesterol_Level (mean (SD)) 0.783
## Glucose_Level (mean (SD)) 0.135
## Neutrophils_Count (mean (SD)) 0.732
## Platelets_Count (mean (SD)) 0.902
## BiomarcadorBasal_A (mean (SD)) 0.334
## BiomarcadorBasal_B (mean (SD)) 0.166# Variables seleccionadas
selected_vars <- c(
"Age", "Sex", "`Chest pain type`", "BP", "Cholesterol",
"`EKG results`", "`Max HR`", "`Exercise angina`",
"`ST depression`", "`Slope of ST`", "`Number of vessels fluro`",
"Thallium", "BiomarcadorAgudo_C", "Symptom_Scale_Admission",
"Killip_Class_24H", "Recovery_Discharge", "`3M_Prognosis`",
"Treatment_Type", "Cholesterol_Level", "Glucose_Level",
"Neutrophils_Count", "Platelets_Count","BiomarcadorBasal_A","BiomarcadorBasal_B")
# Filtrar variables existentes en el conjunto de datos
repre <- intersect(selected_vars, names(data_total))
# Remover la variable objetivo del conjunto `repre`
repre <- repre[repre != "Treatment_Response"]
# Crear la fórmula con las variables seleccionadas
formula <- as.formula(paste("Treatment_Response ~", paste(repre, collapse = " + ")))
# Ejecutar los modelos stepwise
stepwise_model_F <- step(
glm(formula, data = data_total, family = binomial),
direction = "forward",
trace = FALSE
)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredstepwise_model_B <- step(
glm(formula, data = data_total, family = binomial),
direction = "backward",
trace = FALSE
)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurredstepwise_model_BO <- step(
glm(formula, data = data_total, family = binomial),
direction = "both",
trace = FALSE
)## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred
## Warning: glm.fit: fitted probabilities numerically 0 or 1 occurred# Obtener los AICs de los modelos
aic_values <- c(
Forward = AIC(stepwise_model_F),
Backward = AIC(stepwise_model_B),
Both = AIC(stepwise_model_BO)
)
# Encontrar el modelo con el menor AIC
mejor_direccion <- names(which.min(aic_values))
mejor_aic <- min(aic_values)
# Mostrar resultados
cat("El modelo con el menor AIC es:", mejor_direccion, "\n")## El modelo con el menor AIC es: Backwardcat("Valor del AIC:", mejor_aic, "\n")## Valor del AIC: 102.1959# Resumen del mejor modelo
if (mejor_direccion == "Forward") {
resumen_mejor_modelo <- summary(stepwise_model_F)
} else if (mejor_direccion == "Backward") {
resumen_mejor_modelo <- summary(stepwise_model_B)
} else {
resumen_mejor_modelo <- summary(stepwise_model_BO)
}
print(resumen_mejor_modelo)##
## Call:
## glm(formula = Treatment_Response ~ Age + Cholesterol + BiomarcadorAgudo_C +
## Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type +
## Glucose_Level + BiomarcadorBasal_A, family = binomial, data = data_total)
##
## Coefficients:
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) 7.761628 3.409247 2.277 0.022808 *
## Age -0.063286 0.037258 -1.699 0.089399 .
## Cholesterol 0.007890 0.005194 1.519 0.128695
## BiomarcadorAgudo_C 0.046963 0.014221 3.302 0.000959 ***
## Symptom_Scale_Admission -0.059049 0.025192 -2.344 0.019080 *
## Killip_Class_24H -0.774185 0.310719 -2.492 0.012717 *
## Treatment_TypeMedication Only -1.764648 0.860843 -2.050 0.040374 *
## Treatment_TypeNo Treatment -2.045430 1.593997 -1.283 0.199419
## Glucose_Level -0.045618 0.020842 -2.189 0.028617 *
## BiomarcadorBasal_A 0.114551 0.087279 1.312 0.189366
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## (Dispersion parameter for binomial family taken to be 1)
##
## Null deviance: 114.742 on 120 degrees of freedom
## Residual deviance: 82.196 on 111 degrees of freedom
## AIC: 102.2
##
## Number of Fisher Scoring iterations: 7set.seed(123) # Para reproducibilidad
# Dividir datos en conjunto de entrenamiento (70%) y prueba (30%)
sample_index <- sample(1:nrow(data_total), size = 0.8 * nrow(data_total))
train_data <- data_total[sample_index, ]
test_data <- data_total[-sample_index, ]
# Asegurarte de que la variable dependiente sea categórica
train_data$`3M_Prognosis` <- as.factor(train_data$Treatment_Response)
test_data$`3M_Prognosis` <- as.factor(test_data$Treatment_Response)
# Cargar librería RandomForest
library(randomForest)
# Entrenar el modelo Random Forest
rf_model <- randomForest(
Treatment_Response ~ Age + BiomarcadorAgudo_C +
Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type +
Glucose_Level,
data = train_data,
ntree = 500, # Número de árboles
mtry = 3, # Variables consideradas en cada split
importance = TRUE # Calcular importancia de las variables
)
# Resumen del modelo
print(rf_model)##
## Call:
## randomForest(formula = Treatment_Response ~ Age + BiomarcadorAgudo_C + Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type + Glucose_Level, data = train_data, ntree = 500, mtry = 3, importance = TRUE)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 3
##
## OOB estimate of error rate: 45.83%
## Confusion matrix:
## Good Poor Regular class.error
## Good 3 7 9 0.8421053
## Poor 3 36 6 0.2000000
## Regular 5 14 13 0.5937500# Predicciones en el conjunto de prueba
predicciones_rf <- predict(rf_model, newdata = test_data, type = "response")
# Matriz de confusión
matriz_confusion <- table(Predicho = predicciones_rf, Real = test_data$Treatment_Response)
print(matriz_confusion)## Real
## Predicho Good Poor Regular
## Good 2 1 1
## Poor 0 8 3
## Regular 1 1 8# Extraer valores de la matriz de confusión
TN <- matriz_confusion[1,1]
FP <- matriz_confusion[1,2]
FN <- matriz_confusion[2,1]
TP <- matriz_confusion[2,2]
# Sensibilidad (Recall o TPR)
sensitivity <- TP / (TP + FN)
# Especificidad (TNR)
specificity <- TN / (TN + FP)
# PPV (Valor Predictivo Positivo)
ppv <- TP / (TP + FP)
# NPV (Valor Predictivo Negativo)
npv <- TN / (TN + FN)
# Mostrar métricas
cat("Sensibilidad: ", sensitivity, "\n")## Sensibilidad: 1cat("Especificidad: ", specificity, "\n")## Especificidad: 0.6666667cat("PPV: ", ppv, "\n")## PPV: 0.8888889cat("NPV: ", npv, "\n")## NPV: 1# Calcular Accuracy
accuracy <- sum(diag(matriz_confusion)) / sum(matriz_confusion)
cat("Accuracy del modelo:", accuracy, "\n")## Accuracy del modelo: 0.72# Gráfico de importancia de variables
importance <- importance(rf_model)
varImpPlot(rf_model)
# Curva ROC y AUC
library(pROC)
roc_curve <- roc(as.numeric(test_data$`3M_Prognosis`), as.numeric(predicciones_rf))## Warning in roc.default(as.numeric(test_data$`3M_Prognosis`), as.numeric(predicciones_rf)): 'response' has more
## than two levels. Consider setting 'levels' explicitly or using 'multiclass.roc' instead## Setting levels: control = 1, case = 2## Setting direction: controls < casesplot(roc_curve, main = "Curva ROC para Random Forest", col = "blue")
auc <- auc(roc_curve)
cat("Área bajo la curva (AUC):", auc, "\n")## Área bajo la curva (AUC): 0.65test_data$Treatment_Response <- factor(test_data$Treatment_Response)
predicciones_rf <- factor(predicciones_rf, levels = levels(test_data$Treatment_Response))
# Crear matriz de confusión
conf_matrix <- table(Predicho = predicciones_rf, Real = test_data$Treatment_Response)
# Verificar si la matriz de confusión tiene sentido
print(conf_matrix)## Real
## Predicho Good Poor Regular
## Good 2 1 1
## Poor 0 8 3
## Regular 1 1 8# Convertir la matriz en un data frame para ggplot2
conf_df <- as.data.frame(as.table(conf_matrix))
# Graficar usando ggplot2
ggplot(conf_df, aes(x = Predicho, y = Real, fill = Freq)) +
geom_tile(color = "white") +
scale_fill_gradient(low = "lightblue", high = "darkblue") +
geom_text(aes(label = Freq), vjust = 1, color = "white", size = 5) +
labs(title = "Matriz de Confusión",
x = "Predicho", y = "Real") +
theme_minimal()
metrics_summary <- data.frame(
Metric = c("Sensibilidad", "Especificidad", "PPV", "NPV", "Accuracy", "AUC"),
Value = c(sensitivity, specificity, ppv, npv, accuracy, auc)
)
# Mostrar la tabla resumen
print(metrics_summary)# Predicciones en probabilidades
pred_prob <- predict(rf_model, newdata = test_data, type = "prob")
# Ajustar el umbral para la clase positiva (puedes probar diferentes umbrales)
threshold <- 0.7 # Prueba con 0.6, 0.7 o un valor que te convenga
pred_class <- ifelse(pred_prob[,2] > threshold, 1, 0)
# Calcular la frecuencia de las clases
class_weights <- table(train_data$Treatment_Response)
class_weights <- 1 / class_weights
names(class_weights) <- levels(train_data$Treatment_Response)
# Ajustar el modelo Random Forest con pesos de clases
rf_model <- randomForest(
Treatment_Response ~ Age + BiomarcadorAgudo_C +
Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type +
Glucose_Level,
data = train_data,
ntree = 500,
mtry = 4,
importance = TRUE,
classwt = class_weights # Pesos ajustados
)
# Resumen del modelo
print(rf_model)##
## Call:
## randomForest(formula = Treatment_Response ~ Age + BiomarcadorAgudo_C + Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type + Glucose_Level, data = train_data, ntree = 500, mtry = 4, importance = TRUE, classwt = class_weights)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 4
##
## OOB estimate of error rate: 45.83%
## Confusion matrix:
## Good Poor Regular class.error
## Good 4 8 7 0.7894737
## Poor 3 35 7 0.2222222
## Regular 5 14 13 0.5937500# Calcular la frecuencia de las clases
class_weights <- table(train_data$Treatment_Response)
class_weights <- 1 / class_weights
names(class_weights) <- levels(train_data$Treatment_Response)
# Ajustar el modelo Random Forest con pesos de clases
rf_model <- randomForest(
Treatment_Response ~ Age + BiomarcadorAgudo_C +
Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type +
Glucose_Level,
data = train_data,
ntree = 500,
mtry = 4,
importance = TRUE,
classwt = class_weights # Pesos ajustados
)
# Resumen del modelo
print(rf_model)##
## Call:
## randomForest(formula = Treatment_Response ~ Age + BiomarcadorAgudo_C + Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type + Glucose_Level, data = train_data, ntree = 500, mtry = 4, importance = TRUE, classwt = class_weights)
## Type of random forest: classification
## Number of trees: 500
## No. of variables tried at each split: 4
##
## OOB estimate of error rate: 45.83%
## Confusion matrix:
## Good Poor Regular class.error
## Good 4 8 7 0.7894737
## Poor 2 35 8 0.2222222
## Regular 5 14 13 0.5937500# Ajustar el parámetro nodesize
rf_model <- randomForest(
Treatment_Response ~ Age + BiomarcadorAgudo_C +
Symptom_Scale_Admission + Killip_Class_24H + Treatment_Type +
Glucose_Level,
data = train_data,
ntree = 500,
mtry = 4,
importance = TRUE,
nodesize = 10 # Cambiar este valor según sea necesario
)
conf_matrix <- table(Predicted = pred_class, Actual = test_data$Treatment_Response)
# Calcular especificidad
TN <- conf_matrix[1, 1]
FP <- conf_matrix[1, 2]
specificity <- TN / (TN + FP)
cat("Especificidad:", specificity, "\n")## Especificidad: 0.375# Calcular AUC y otras métricas (si es necesario)
library(pROC)
roc_curve <- roc(test_data$Treatment_Response, pred_prob[, 2])## Warning in roc.default(test_data$Treatment_Response, pred_prob[, 2]): 'response' has more than two levels.
## Consider setting 'levels' explicitly or using 'multiclass.roc' instead## Setting levels: control = Good, case = Poor## Setting direction: controls < casesauc_value <- auc(roc_curve)
cat("AUC:", auc_value, "\n")## AUC: 1PARTE 2,2: analisis de supervivencia engun la respuesta al tratamiento.
library(survival)
library(survminer)
# Asegúrate de que las fechas están en formato Date
data_total$Admission_Date <- as.Date(data_total$Admission_Date)
data_total$Death_Date <- as.Date(data_total$Death_Date)
# Crear el objeto de supervivencia
# Usamos la diferencia entre 'Admission_Date' y 'Death_Date' como tiempo de seguimiento
# Si Death_Date es NA, el evento no ocurrió (censura)
surv_object <- Surv(time = as.numeric(difftime(data_total$Death_Date, data_total$Admission_Date, units = "days")),
event = ifelse(!is.na(data_total$Death_Date), 1, 0))
# Ajustar el modelo Kaplan-Meier estratificado por respuesta al tratamiento
km_fit <- survfit(surv_object ~ Treatment_Response, data = data_total)
# Graficar las curvas de supervivencia
ggsurvplot(
km_fit,
data = data_total,
conf.int = TRUE, # Mostrar intervalos de confianza
pval = TRUE, # Mostrar valor p
risk.table = TRUE, # Agregar tabla de riesgos
title = "Curvas de Supervivencia según Respuesta al Tratamiento",
xlab = "Días",
ylab = "Probabilidad de Supervivencia",
legend.title = "Respuesta al Tratamiento",
palette = c("#1b9e77", "#d95f02", "#7570b3") # Colores para las categorías
)