1 Context

According to the World Health Organization (WHO) stroke is the 2nd leading cause of death globally, responsible for approximately 11% of total deaths.

This dataset is used to predict whether a patient is likely to get stroke based on the input parameters like gender, age, various diseases, and smoking status. Each row in the data provides relavant information about the patient.

2 Attribute Information

id: unique identifier
gender: “Male”, “Female” or “Other”
age: age of the patient
hypertension: 0 if the patient doesn’t have hypertension, 1 if the patient has hypertension
heart_disease: 0 if the patient doesn’t have any heart diseases, 1 if the patient has a heart disease
ever_married: “No” or “Yes”
work_type: “children”, “Govt_jov”, “Never_worked”, “Private” or “Self-employed”
Residence_type: “Rural” or “Urban”
avg_glucose_level: average glucose level in blood
bmi: body mass index
smoking_status: “formerly smoked”, “never smoked”, “smokes” or “Unknown”*
stroke: 1 if the patient had a stroke or 0 if not *Note: “Unknown” in smoking_status means that the information is unavailable for this patient

3 Source

https://www.kaggle.com/datasets/fedesoriano/stroke-prediction-dataset

library(dplyr)

4 Data Transformation

# Read Data
stroke <- read.csv(file = "strokedata.csv", stringsAsFactors = T)

# Check Data
glimpse(stroke)

#> Rows: 5,110
#> Columns: 12
#> $ id                <int> 9046, 51676, 31112, 60182, 1665, 56669, 53882, 10434~
#> $ gender            <fct> Male, Female, Male, Female, Female, Male, Male, Fema~
#> $ age               <dbl> 67, 61, 80, 49, 79, 81, 74, 69, 59, 78, 81, 61, 54, ~
#> $ hypertension      <int> 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1~
#> $ heart_disease     <int> 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0~
#> $ ever_married      <fct> Yes, Yes, Yes, Yes, Yes, Yes, Yes, No, Yes, Yes, Yes~
#> $ work_type         <fct> Private, Self-employed, Private, Private, Self-emplo~
#> $ Residence_type    <fct> Urban, Rural, Rural, Urban, Rural, Urban, Rural, Urb~
#> $ avg_glucose_level <dbl> 228.69, 202.21, 105.92, 171.23, 174.12, 186.21, 70.0~
#> $ bmi               <fct> 36.6, N/A, 32.5, 34.4, 24, 29, 27.4, 22.8, N/A, 24.2~
#> $ smoking_status    <fct> formerly smoked, never smoked, never smoked, smokes,~
#> $ stroke            <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1~

# Data Wrangling

stroke_clean <- mutate(.data = stroke,
                       bmi = as.character(bmi),
                       id = as.character(id))

glimpse(stroke_clean)

#> Rows: 5,110
#> Columns: 12
#> $ id                <chr> "9046", "51676", "31112", "60182", "1665", "56669", ~
#> $ gender            <fct> Male, Female, Male, Female, Female, Male, Male, Fema~
#> $ age               <dbl> 67, 61, 80, 49, 79, 81, 74, 69, 59, 78, 81, 61, 54, ~
#> $ hypertension      <int> 0, 0, 0, 0, 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1~
#> $ heart_disease     <int> 1, 0, 1, 0, 0, 0, 1, 0, 0, 0, 0, 1, 0, 1, 1, 0, 1, 0~
#> $ ever_married      <fct> Yes, Yes, Yes, Yes, Yes, Yes, Yes, No, Yes, Yes, Yes~
#> $ work_type         <fct> Private, Self-employed, Private, Private, Self-emplo~
#> $ Residence_type    <fct> Urban, Rural, Rural, Urban, Rural, Urban, Rural, Urb~
#> $ avg_glucose_level <dbl> 228.69, 202.21, 105.92, 171.23, 174.12, 186.21, 70.0~
#> $ bmi               <chr> "36.6", "N/A", "32.5", "34.4", "24", "29", "27.4", "~
#> $ smoking_status    <fct> formerly smoked, never smoked, never smoked, smokes,~
#> $ stroke            <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1~

In 2015, the World Health Organization under the United Nations officially revised the age standards. According to the new age classification (in years old):

children: 0-12
adolescent/youth: 13-24
young age: 25-44
middle age: 45-60
elderly age: 61-75
senile age: 76-90
long-livers: after 90

# Tambah kolom klasifikasi usia (age_category)

x <- c("0-12 (children)", 
       "13-24 (adolescent/youth)",
       "25-44 (young age)", 
       "45-60 (middle age)", 
       "61-75 (elderly age)",
       "76-90 (senile age)", 
       ">90 (long-livers)"
       ) 

stroke_clean <- stroke_clean %>% 
mutate(age_category = if_else(age >= 0 & age <= 12, "0-12 (children)",
                if_else(age >= 13 & age <= 24, "13-24 (adolescent/youth)",  
                if_else(age >= 25 & age <= 44, "25-44 (young age)",
                if_else(age >= 45 & age <= 60, "45-60 (middle age)",
                if_else(age >= 61 & age <= 75, "61-75 (elderly age)",
                if_else(age >= 76 & age  <= 90, "76-90 (senile age)",
                if_else(age > 90, ">90 (long-livers)", "")))))))) %>%
    mutate(age_category = factor(age_category, levels = x))

head(stroke_clean)

# Check Data 
anyNA(stroke_clean)

#> [1] FALSE

colSums(is.na(stroke_clean))

#>                id            gender               age      hypertension 
#>                 0                 0                 0                 0 
#>     heart_disease      ever_married         work_type    Residence_type 
#>                 0                 0                 0                 0 
#> avg_glucose_level               bmi    smoking_status            stroke 
#>                 0                 0                 0                 0 
#>      age_category 
#>                 0

5 Business Questions:

Kelompok usia mana yang paling rentan mengalami stroke?
Kelompok gender mana yang rentan mengalami stroke?
Kelompok mana yang paling rentan mengalami stroke berdasarkan kategori usia, lokasi tinggal, riwayat merokok, menikah, dan pengalaman kerja?
Buat plot interaktif dari kasus stroke terhadap gender, area tempat tinggal, kelompok usia, dan riwayat merokok

5.1 Kelompok usia mana yang paling rentan mengalami stroke?

stroke_age <- stroke_clean %>% 
  filter(stroke==1) %>% 
  group_by(age_category) %>% 
  summarise(count=n()) %>% 
  ungroup() %>% 
  arrange(-count)

head(stroke_age)

5.1.1 Kesimpulan:

Kelompok usia “senile age” (76-90 thn) paling rentan mengalami stroke

5.2 Kelompok gender mana yang rentan mengalami stroke?

stroke_gender <- stroke_clean %>% 
  filter(stroke==1) %>% 
  group_by(gender) %>% 
  summarise(count=n()) %>% 
  ungroup() %>% 
  arrange(gender)

head(stroke_gender)

5.2.1 Kesimpulan:

Gender perempuan paling rentan mengalami stroke

5.3 Kelompok mana yang paling rentan mengalami stroke berdasarkan kategori usia, lokasi tinggal, riwayat merokok, menikah, dan pengalaman kerja?

stroke_most <- stroke_clean %>% 
  filter(stroke==1) %>% 
  group_by(age_category,gender,Residence_type,smoking_status, ever_married, work_type) %>% 
  summarise(count=n()) %>% 
  ungroup() %>% 
  arrange(-count)

head (stroke_most)

5.3.1 Kesimpulan:

Kelompok perempuan dengan pengalaman bekerja sebagai karyawan swasta, tinggal di area pedesaan, berkeluarga, tidak pernah merokok, usia 75-90 tahun

5.4 Buat plot interaktif dari kasus stroke terhadap kelompok usia, riwayat merokok, gender, dan area tempat tinggal

library(ggplot2) 
library(plotly) 
library(glue) 
library(scales)

5.4.1 Tingkat Kejadian Stroke berdasarkan Gender

stroke_gender_pie <- data.frame("Gender"=rownames(stroke_gender), stroke_gender)

data <- stroke_gender[,c('gender','count')]

colors <- c('rgb(0,0,0)', 'rgb(184,184,184)')

fig <- plot_ly(data, labels = ~gender, values = ~count, type = 'pie',
               textposition = 'inside',
               textinfo = 'label+percent',
               insidetextfont = list(color = '#FFFFFF'),
               hoverinfo = 'text',
               text = ~paste(count, 'stroke cases'),
               marker = list(colors = colors,
                      line = list(color = '#FFFFFF', width = 1),
               showlegend = FALSE))

fig <- fig %>% layout(title = 'Stroke Cases by Gender',
                      xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
                      yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))

fig

5.4.1.1 Insight:

Kasus stroke terjadi hampir sama banyak pada laki-laki dan perempuan, dimana tingkat kejadian pada kaum perempuan agak lebih tinggi daripada laki-laki.

5.4.2 Tingkat Kejadian Stroke berdasarkan Area Tempat Tinggal

## buat pie chart nya

stroke_res2 <- data.frame("Residence Area"=rownames(stroke_res), stroke_res)

data2 <- stroke_res[,c('Residence_type','count')]

colors2 <- c('rgb(0,0,0)', 'rgb(184,184,184)')

fig2 <- plot_ly(data2, labels = ~Residence_type, values = ~count, type = 'pie',
               textposition = 'inside',
               textinfo = 'label+percent',
               insidetextfont = list(color = '#FFFFFF'),
               hoverinfo = 'text',
               text = ~paste(count, 'stroke cases'),
               marker = list(colors = colors2,
                      line = list(color = '#FFFFFF', width = 1),
               showlegend = FALSE))

fig2 <- fig2 %>% layout(title = 'Stroke Cases by Residence Area',
                      xaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE),
                      yaxis = list(showgrid = FALSE, zeroline = FALSE, showticklabels = FALSE))

fig2

5.4.2.1 Insight:

Kasus stroke juga hampir sama banyak terjadi baik bagi mereka yang tinggal di area perkotaan maupun pedesaan, dimana tingkat kejadian pada penduduk kota agak lebih tinggi daripada mereka yang tingal di area pedesaan.

5.4.3 Tingkat Kejadian Stroke berdasarkan Kelompok Usia

stroke_most_age <- stroke_clean %>% 
  filter(stroke==1) %>% 
  group_by(age_category) %>% 
  summarise(count=n()) %>% 
  ungroup() %>% 
  arrange(-count)

stroke_most_age2 <- stroke_most_age %>% 
  mutate(label = glue(
    "Age Category: {age_category}
    Stroke Cases: {count}"
  ))

plot1 <- ggplot(data = stroke_most_age2, aes(x = count,
                                             y = reorder(age_category, count),
                                             text = label)) +
  geom_col(aes(fill = count)) +
  scale_fill_gradient(low="grey", high="black") +
  labs(title = "Stroke Cases by Age Category",
       x = "Stroke Cases",
       y = NULL) +
  theme_minimal() +
  theme(legend.position = "none")

ggplotly(plot1, tooltip = "text")

5.4.3.1 Insight:

Pareto kasus stroke paling banyak terjadi pada kelompok usia lansia (>60 tahun), meskipun ada beberapa kasus stroke juga yang terjadi pada usia produktif (25-60 thn).

5.4.4 Tingkat Kejadian Stroke berdasarkan Riwayat Merokok

stroke_smoke <- stroke_clean %>% 
  filter(stroke==1) %>% 
  group_by(smoking_status) %>% 
  summarise(count=n()) %>% 
  ungroup() %>% 
  arrange(-count)

stroke_smoke2 <- stroke_smoke %>% 
  mutate(label = glue(
    "Smoking History: {smoking_status}
    Stroke Cases: {count}"
  ))

plot2 <- ggplot(data = stroke_smoke2, aes(x = count,
                                             y = reorder(smoking_status, count),
                                             text = label)) +
  geom_col(aes(fill = count)) +
  scale_fill_gradient(low="grey", high="black") +
  labs(title = "Stroke Cases by Smoking History",
       x = "Stroke Cases",
       y = NULL) +
  theme_minimal() +
  theme(legend.position = "none")

ggplotly(plot2, tooltip = "text")

5.4.4.1 Insight:

Temuan menarik bahwa kasus stroke paling banyak terjadi pada mereka yang tidak pernah merokok. Perlu studi literasi lebih jauh mengenai penyebab kondisi ini.

LBB_IP: Stroke Cases

o_diella

12 May, 2022