🏥 Heart Disease Risk Prediction — Machine Learning

---

📌 Project Overview

A complete Machine Learning pipeline for predicting heart disease risk in patients using clinical and lifestyle data. The project covers the full data science workflow: data generation, EDA, preprocessing, model training, evaluation, and interpretation.

Three classification models are compared:

• 📊 Logistic Regression
• 🌲 Random Forest ← Best Model (AUC = 1.000)
• 🚀 Gradient Boosting (AUC = 0.955)

---

🎯 Objective

Predict whether a patient is at risk of heart disease based on 10 clinical and lifestyle features, enabling early intervention and preventive care.

---

📊 Dashboard Preview

Exploratory Data Analysis

ML Model Evaluation

---

📁 Project Structure

Heart-Disease-Prediction/
│
├── 📄 healthcare_data.csv       # Patient dataset (2,000 records)
├── 🐍 generate_data.py          # Synthetic data generator
├── 🐍 ml_analysis.py            # EDA + ML training & evaluation
├── 📊 eda_dashboard.png         # Exploratory analysis charts
├── 📊 ml_results.png            # Model comparison & ROC curves
└── 📖 README.md

---

📦 Dataset Description

Feature	Type	Description
Age	Integer	Patient age (20–80)
BMI	Float	Body Mass Index
Glucose	Integer	Blood glucose level (mg/dL)
Blood_Pressure	Integer	Systolic blood pressure
Cholesterol	Integer	Total cholesterol (mg/dL)
Smoking	Binary	1 = Smoker, 0 = Non-smoker
Exercise_Level	Ordinal	0=None, 1=Low, 2=Medium, 3=High
Family_History	Binary	1 = Family history of heart disease
Stress_Level	Integer	Self-reported stress (1–10)
Sleep_Hours	Float	Average daily sleep (hours)
Disease	Binary	Target: 1 = Disease, 0 = Healthy

---

🔬 ML Pipeline

Raw Data
   │
   ▼
Exploratory Data Analysis (EDA)
   │  ├─ Distribution plots
   │  ├─ Correlation heatmap
   │  └─ Risk factor analysis
   │
   ▼
Preprocessing
   │  ├─ Handle class imbalance (oversampling)
   │  └─ Feature scaling (StandardScaler)
   │
   ▼
Model Training (3 models)
   │  ├─ Logistic Regression
   │  ├─ Random Forest (200 estimators)
   │  └─ Gradient Boosting (200 estimators)
   │
   ▼
Evaluation
   │  ├─ ROC-AUC Score
   │  ├─ Confusion Matrix
   │  ├─ Precision / Recall / F1
   │  └─ Feature Importance
   │
   ▼
Best Model: Random Forest (AUC = 1.000)

---

📈 Model Performance

Model	AUC Score	Notes
Logistic Regression	0.612	Baseline linear model
Gradient Boosting	0.955	Strong ensemble model
Random Forest	1.000	✅ Best — recommended for deployment

---

🔍 Key Findings

• 🚬 Smoking is the strongest risk factor for disease
• 👨‍👩‍👦 Family History doubles disease probability
• 🏃 Regular exercise significantly reduces risk
• 🍬 High glucose & BMI are correlated with disease
• 😴 Poor sleep (< 5 hours) increases risk by ~40%
• 💪 Random Forest outperforms other models with perfect AUC on test set

---

🛠️ Technologies Used

pandas         # Data manipulation
numpy          # Numerical operations
matplotlib     # Custom visualizations
seaborn        # Statistical plots
scikit-learn   # ML models, metrics, preprocessing

---

▶️ How to Run

# 1. Clone the repository
git clone https://github.com/eng-heba-tech/Heart-Disease-Prediction
cd Heart-Disease-Prediction

# 2. Install dependencies
pip install pandas numpy matplotlib seaborn scikit-learn

# 3. Generate dataset
python generate_data.py

# 4. Run full analysis + ML
python ml_analysis.py

---

🧠 Sample Code — Model Training

from sklearn.ensemble import RandomForestClassifier
from sklearn.model_selection import train_test_split
from sklearn.metrics import roc_auc_score

X_train, X_test, y_train, y_test = train_test_split(
    X, y, test_size=0.2, random_state=42, stratify=y)

rf = RandomForestClassifier(n_estimators=200, random_state=42)
rf.fit(X_train, y_train)

auc = roc_auc_score(y_test, rf.predict_proba(X_test)[:,1])
print(f"Random Forest AUC: {auc:.3f}")  # Output: 1.000

---

👩‍💻 Author

Eng. Heba | Data Analyst & ML Engineer

• 🐙 GitHub: @eng-heba-tech
• 📧 eng.heba.tech@gmail.com
• 🌍 Egypt

---

⭐ If you find this project useful, please give it a star!