🛡️ Heart Disease_Prediction Model

A machine learning project to predict the presence of heart disease based on clinical and physiological health parameters. It helps in early diagnosis, risk identification, and efficient medical decision-making.

---

🎯 Objective

To build a predictive ML model that determines whether a person is likely to have heart disease using features like age, cholesterol level, blood pressure, chest pain type, heart rate, and more.

For Doctors & Healthcare Teams

• Assess heart disease risk quickly

• Support clinical decision-making

• Identify high-risk patients for early treatment

• Prioritize preventive care in critical cases

For Individuals

• Understand personal risk based on health metrics

• Learn which factors affect heart-related problems

• Encourage lifestyle changes (BMI, BP, exercise, etc.)

For Healthcare Systems

• Predict population-level heart disease trends

• Enable better planning of medical resources

• Reduce emergency risks through early detection

---

🌟 Attributes in this Model

Attribute	Description	Effect
Age	Age of the person	Older age → higher risk
Sex	Male/Female	Males have slightly higher risk
Chest Pain Type (cp)	4 categories of pain	Certain types strongly indicate disease
Trestbps	Resting blood pressure	Higher BP → potential heart issues
Chol	Serum cholesterol	High levels increase risk
Fbs	Fasting blood sugar	High sugar → heart complications
Restecg	Resting ECG results	Abnormal ECG → higher risk
Thalach	Maximum heart rate	Lower heart rate → higher risk
Exang	Exercise-induced angina	Positive angina → higher chance of disease
Oldpeak	ST depression	Higher values show abnormal heart stress
Slope	Slope of peak exercise ST	Indicates heart performance
Ca	Number of major vessels	More vessels blocked → higher risk
Thal	Thalassemia test result	Abnormal result → high correlation with disease

---

📈 Project Workflow

├──Data Loading
├── Data Cleaning and Preprocessing
├── Exploratory Data Analysis (EDA)
├── Feature Selection
├── Train–Test Split
├── Model Training (Logistic Regression)
├── Model Evaluation (Accuracy, ROC Curve, Confusion Matrix)
├── Prediction on New Patient Data
└── Future Suggestions and Improvements

⚙️ Tools & Libraries

• Python

• NumPy

• Pandas

• Matplotlib

• Scikit-learn

• Jupyter Notebook / VS Code

📉 Model Evaluation Metrics

1. Accuracy Score → % of correct predictions

2. Confusion Matrix → TP, TN, FP, FN

3. Precision & Recall → critical for medical domain

4. ROC–AUC Score → how well the model separates classes

Higher values indicate better performance in identifying at-risk patients

🧠 Why Logistic Regression?

• Ideal for binary classification (disease/no disease)

• Simple, interpretable, and medically explainable

• Lightweight and performs well on structured datasets

• Provides clear probability output

• Strong baseline before using advanced models

🔍 Key Insights

• Chest pain type is one of the strongest indicators

• Maximum heart rate (thalach) strongly correlates with target

• Higher oldpeak values show abnormal stress behavior

• Males show slightly higher cases than females

• Cholesterol and BP vary but are not the strongest predictors

---

👩‍💻 Author

Khushi Goyal

Machine Learning & Web Development Enthusiast