Background

Heart disease is a leading global health concern, accounting for a significant number of deaths each year. It encompasses various conditions affecting the heart, including coronary artery disease, heart attacks, and arrhythmias. Early detection and accurate diagnosis are critical in preventing complications, reducing healthcare costs, and improving patient outcomes.

Hospitals, healthcare providers, and insurance companies are increasingly interested in using data-driven insights to support early intervention strategies. Early identification of individuals at risk of developing heart disease not only improves patient outcomes but also plays a vital role in preventive cardiology initiatives.

The Heart Disease dataset from the UCI Machine Learning Repository provides a valuable collection of patient attributes, including demographic information, medical test results, and lifestyle factors. Key indicators such as age, cholesterol levels, blood pressure, and chest pain type are included, offering a rich basis for analysis. Analyzing this data can help uncover patterns and correlations that contribute to heart disease and support better decision-making in medical settings.

Data visualization plays a crucial role in understanding complex medical datasets. By employing exploratory data analysis (EDA) and visualization techniques, we can uncover hidden trends, compare different risk factors, and generate insights that support clinical and operational decisions. This project focuses on utilizing various data visualization methods to analyze the UCI Heart Disease dataset and provide a clearer understanding of the factors influencing heart disease risk.

Business Objectives

• Primary Objective: Gain exploratory insights into key factors associated with heart disease from the UCI dataset through data visualization and analysis.
• Secondary Objective: Identify patterns, trends, and relationships within the dataset that can later guide predictive modeling efforts and hypothesis generation.
• Long-Term Vision: Equip healthcare professionals, researchers, and organizations with better data insights that can lead to improved decision-making, early intervention strategies, and preventive healthcare initiatives in cardiology.

At this stage, no predictive model is being developed. The goal is purely exploratory: to generate hypotheses, guide future analytical strategies, and understand the structure and relationships in the data.

Data Collection

The dataset used is the Heart Disease dataset from the UCI Machine Learning Repository. It is a compilation of four databases (Cleveland, Hungary, Switzerland, and Long Beach VA), but most studies (and this project) focus primarily on the Cleveland dataset, which has the most complete records.

The dataset contains various features related to patients’ health and demographic information. We will explore the dataset to understand its structure and relationships between variables.

Data Description

The dataset contains 14 key attributes that are either numerical or categorical. These attributes are:

1. age: Age of the patient (numeric)
2. sex: Gender of the patient (1 = male, 0 = female)
3. cp: Chest pain type (categorical: 1-4)
4. trestbps: Resting blood pressure (numeric)
5. chol: Serum cholesterol (numeric)
6. fbs: Fasting blood sugar (1 = true, 0 = false)
7. restecg: Resting electrocardiographic results (categorical)
8. thalach: Maximum heart rate achieved (numeric)
9. exang: Exercise-induced angina (1 = yes, 0 = no)
10. oldpeak: ST depression induced by exercise (numeric)
11. slope: The slope of the peak exercise ST segment (categorical)
12. ca: Number of major vessels (0-3, numeric)
13. thal: Thalassemia (categorical: 1 = normal, 2 = fixed defect, 3 = reversible defect)
14. target: Heart disease (1 = disease, 0 = no disease)

Note: Some fields (e.g., “ca” and “thal”) may contain missing or invalid values coded as ‘?’, which need attention during preprocessing.

Data Dictionary

The dataset contains 303 instances (patients) and 14 main attributes plus the target variable (presence or absence of heart disease).

Initial Data Quality Assessment

• Missing Values: Some fields like ca and thal may have missing values or unknown entries (‘?’).
• Data Types: Some categorical variables are encoded numerically and will need to be interpreted correctly during analysis.
• Class Imbalance: Preliminary checks suggest the dataset is relatively balanced between presence and absence of disease, but this will be verified.
• Outliers: Numerical fields such as chol (cholesterol) and trestbps (blood pressure) may have outliers that need to be detected and considered in analysis.

Data Loading

# Create url object to retrieve the dataset from UCI Machine Learning Repository
url <- "https://archive.ics.uci.edu/ml/machine-learning-databases/heart-disease/processed.cleveland.data"

# Read the dataset into a dataframe
Heart.df <- read.csv(url(url), header = FALSE, na.strings = "?")

dim(Heart.df)

[1] 303  14

head(Heart.df)

  V1 V2 V3  V4  V5 V6 V7  V8 V9 V10 V11 V12 V13 V14
1 63  1  1 145 233  1  2 150  0 2.3   3   0   6   0
2 67  1  4 160 286  0  2 108  1 1.5   2   3   3   2
3 67  1  4 120 229  0  2 129  1 2.6   2   2   7   1
4 37  1  3 130 250  0  0 187  0 3.5   3   0   3   0
5 41  0  2 130 204  0  2 172  0 1.4   1   0   3   0
6 56  1  2 120 236  0  0 178  0 0.8   1   0   3   0

Data Preprocessing

colnames(Heart.df) <- c("age", "sex", "cp", "trestbps", "chol", "fbs", "restecg", "thalach", "exang", "oldpeak", "slope", "ca", "thal", "target")

str(Heart.df)

'data.frame':   303 obs. of  14 variables:
 $ age     : num  63 67 67 37 41 56 62 57 63 53 ...
 $ sex     : num  1 1 1 1 0 1 0 0 1 1 ...
 $ cp      : num  1 4 4 3 2 2 4 4 4 4 ...
 $ trestbps: num  145 160 120 130 130 120 140 120 130 140 ...
 $ chol    : num  233 286 229 250 204 236 268 354 254 203 ...
 $ fbs     : num  1 0 0 0 0 0 0 0 0 1 ...
 $ restecg : num  2 2 2 0 2 0 2 0 2 2 ...
 $ thalach : num  150 108 129 187 172 178 160 163 147 155 ...
 $ exang   : num  0 1 1 0 0 0 0 1 0 1 ...
 $ oldpeak : num  2.3 1.5 2.6 3.5 1.4 0.8 3.6 0.6 1.4 3.1 ...
 $ slope   : num  3 2 2 3 1 1 3 1 2 3 ...
 $ ca      : num  0 3 2 0 0 0 2 0 1 0 ...
 $ thal    : num  6 3 7 3 3 3 3 3 7 7 ...
 $ target  : int  0 2 1 0 0 0 3 0 2 1 ...

summary(Heart.df)

      age             sex               cp           trestbps    
 Min.   :29.00   Min.   :0.0000   Min.   :1.000   Min.   : 94.0  
 1st Qu.:48.00   1st Qu.:0.0000   1st Qu.:3.000   1st Qu.:120.0  
 Median :56.00   Median :1.0000   Median :3.000   Median :130.0  
 Mean   :54.44   Mean   :0.6799   Mean   :3.158   Mean   :131.7  
 3rd Qu.:61.00   3rd Qu.:1.0000   3rd Qu.:4.000   3rd Qu.:140.0  
 Max.   :77.00   Max.   :1.0000   Max.   :4.000   Max.   :200.0  
                                                                 
      chol            fbs            restecg          thalach     
 Min.   :126.0   Min.   :0.0000   Min.   :0.0000   Min.   : 71.0  
 1st Qu.:211.0   1st Qu.:0.0000   1st Qu.:0.0000   1st Qu.:133.5  
 Median :241.0   Median :0.0000   Median :1.0000   Median :153.0  
 Mean   :246.7   Mean   :0.1485   Mean   :0.9901   Mean   :149.6  
 3rd Qu.:275.0   3rd Qu.:0.0000   3rd Qu.:2.0000   3rd Qu.:166.0  
 Max.   :564.0   Max.   :1.0000   Max.   :2.0000   Max.   :202.0  
                                                                  
     exang           oldpeak         slope             ca        
 Min.   :0.0000   Min.   :0.00   Min.   :1.000   Min.   :0.0000  
 1st Qu.:0.0000   1st Qu.:0.00   1st Qu.:1.000   1st Qu.:0.0000  
 Median :0.0000   Median :0.80   Median :2.000   Median :0.0000  
 Mean   :0.3267   Mean   :1.04   Mean   :1.601   Mean   :0.6722  
 3rd Qu.:1.0000   3rd Qu.:1.60   3rd Qu.:2.000   3rd Qu.:1.0000  
 Max.   :1.0000   Max.   :6.20   Max.   :3.000   Max.   :3.0000  
                                                 NA's   :4       
      thal           target      
 Min.   :3.000   Min.   :0.0000  
 1st Qu.:3.000   1st Qu.:0.0000  
 Median :3.000   Median :0.0000  
 Mean   :4.734   Mean   :0.9373  
 3rd Qu.:7.000   3rd Qu.:2.0000  
 Max.   :7.000   Max.   :4.0000  
 NA's   :2                       

Heart.df$sex <- ifelse(Heart.df$sex > 0, 1, 0)
Heart.df$fbs <- ifelse(Heart.df$fbs > 0, 1, 0)
Heart.df$exang <- ifelse(Heart.df$exang > 0, 1, 0)
Heart.df$target <- ifelse(Heart.df$target > 0, 1, 0)

sapply(Heart.df, function(x) sum(is.na(x)))

     age      sex       cp trestbps     chol      fbs  restecg  thalach 
       0        0        0        0        0        0        0        0 
   exang  oldpeak    slope       ca     thal   target 
       0        0        0        4        2        0 

# If missing values exist in 'ca' or 'thal', handle them using mean/mode imputation
Heart.df$ca[is.na(Heart.df$ca)] <- median(Heart.df$ca, na.rm = TRUE)
Heart.df$ca[Heart.df$ca == "?"] <- median(Heart.df$ca, na.rm = TRUE)
Heart.df$thal[is.na(Heart.df$thal)] <- median(Heart.df$thal, na.rm = TRUE)
Heart.df$thal[Heart.df$thal == "?"] <- median(Heart.df$ca, na.rm = TRUE)

dupes <- Heart.df[duplicated(Heart.df) | duplicated(Heart.df, fromLast = TRUE), ]
# Print or inspect the duplicate entries
print(dupes)

 [1] age      sex      cp       trestbps chol     fbs      restecg  thalach 
 [9] exang    oldpeak  slope    ca       thal     target  
<0 rows> (or 0-length row.names)

# Define a list of categorical columns with their levels and labels
categorical_columns <- list(
  sex = list(levels = c(0, 1), labels = c("Female", "Male")),
  cp = list(levels = c(1, 2, 3, 4), labels = c("Typical Angina", "Atypical Angina", "Non-Angina", "Asymptomatic")),
  fbs = list(levels = c(0, 1), labels = c("False", "True")),
  restecg = list(levels = c(0, 1, 2), labels = c("Normal", "Wave-abnormality", "Probable")),
  exang = list(levels = c(0, 1), labels = c("No", "Yes")),
  slope = list(levels = c(1, 2, 3), labels = c("Upsloping", "Flat", "Downsloping")),
  thal = list(levels = c(3, 6, 7), labels = c("Normal", "Fixed Defect", "Reversible")),
  target = list(levels = c(1, 0), labels = c("Yes", "No"))
)

# Apply the factor transformation using a loop
for (col in names(categorical_columns)) {
  Heart.df[[col]] <- factor(Heart.df[[col]], 
                            levels = categorical_columns[[col]]$levels, 
                            labels = categorical_columns[[col]]$labels)
}

Handle Outliers

Heart.df <- Heart.df[Heart.df$age > 40 &
                       Heart.df$trestbps < 170 & 
                       Heart.df$chol < 340 & 
                       Heart.df$chol > 150 &
                       Heart.df$thalach > 115 &
                       Heart.df$oldpeak < 2.4, ]

Custom Functions

HeartDiseaseBoxplot <- function(var1, var2) {
  ggplot(Heart.df, aes(x = .data[[var1]],
                       y = .data[[var2]],
                       fill = .data[[var1]])) +
    geom_boxplot() + theme_test() +
    labs(title = paste("Boxplot of", var2, "by", var1),
         x = var1, y = var2, fill = "Heart Disease")
}

HeartDiseaseBar <- function(var) {
  ggplot(Heart.df, aes(x = .data[[var]], fill = target)) +
    geom_bar(position = "dodge") + theme_test() +
    labs(title = paste("Distribution of Heart Disease by", var),
         x = var, fill = "Heart Disease")
}

HeartDiseaseHist <- function(var1) {
  ggplot(Heart.df, aes(x = .data[[var1]], fill = target)) +
    geom_histogram(bins = 15) + theme_test() +
    labs(title = paste("Distribution of", var1),
         x = var1, fill = "Heart Disease")
}

HeartDiseaseScatter <- function(point1, point2){
  ggplot(Heart.df, aes(x = .data[[point1]],
                       y = .data[[point2]],
                       color = target)) +
    geom_point(size = 2) + theme_test() +
    geom_smooth(method = "lm", se = FALSE, color = "blue", formula = y ~ x) +
    labs(title = paste("Scatterplot of", point1, "by", point2),
       x = point1, y = point2, color = "Heart Disease")
}

Boxplots

# Create the plots
p1 <- HeartDiseaseBoxplot("target", "age")
p2 <- HeartDiseaseBoxplot("target", "trestbps")
p3 <- HeartDiseaseBoxplot("target", "chol")
p4 <- HeartDiseaseBoxplot("target", "thalach")
p5 <- HeartDiseaseBoxplot("target", "oldpeak")

# Combine plot using patchwork
(p1 | p2) /
(p3 | p4) /
(p5)

Barplots

# Create the plots
g1 <- ggplot(Heart.df, aes(x=target, fill=target))+
  geom_bar() + theme_test() +
  ggtitle("Distribution of Heart Disease") +
  labs(x = "Heart Disease", fill = "Heart Disease")
g2 <- HeartDiseaseBar("sex")
g3 <- HeartDiseaseBar("cp")
g4 <- HeartDiseaseBar("fbs")
g5 <- HeartDiseaseBar("restecg")
g6 <- HeartDiseaseBar("exang")
g7 <- HeartDiseaseBar("slope")
g8 <- HeartDiseaseBar("thal")

# Combine plot using patchwork
(g1 | g2) /
(g3 | g4) /
(g5 | g6) /
(g7 | g8)

Histogram Distributions

# Create the plots
p1 <- HeartDiseaseHist("age")
p2 <- HeartDiseaseHist("trestbps")
p3 <- HeartDiseaseHist("chol")
p4 <- HeartDiseaseHist("thalach")
p5 <- HeartDiseaseHist("oldpeak")

# Combine plot using patchwork
(p1) /
(p2 | p3) /
(p4 | p5)

Scaterplots

# Create the plots
p1 <- HeartDiseaseScatter("age", "oldpeak")
p2 <- HeartDiseaseScatter("age", "chol")
p3 <- HeartDiseaseScatter("age", "trestbps")
p4 <- HeartDiseaseScatter("age", "thalach")
p5 <- HeartDiseaseScatter("chol", "thalach")
p6 <- HeartDiseaseScatter("trestbps", "chol")
p7 <- HeartDiseaseScatter("thalach", "oldpeak")

# Combine plot using patchwork
(p1 | p2) /
(p3 | p4) /
(p5 | p6) /
(p7)

Pair Plots

# Create a colored pair plot for selected variables
ggpairs(Heart.df[, c("age", "trestbps", "chol", 
                     "thalach", "oldpeak", "target")], 
        aes(color = target, fill = target))

Correlation Matrix

# Selecting only continuous variables
continuous_vars <- c("age", "trestbps", "chol", "thalach", "oldpeak")
continuous_data <- Heart.df %>% select(all_of(continuous_vars))

# Calculating correlation matrix
correlation_matrix <- cor(continuous_data)

# Plotting the correlation matrix
corrplot(correlation_matrix, method = "circle",
         type = "lower", tl.col = "black")