Task 03: Heart Disease Prediction & Analysis

📋 Project Overview

This project was completed as part of my DHC Internship. The goal was to analyze the UCI Heart Disease dataset to identify key health trends and build a machine learning model capable of predicting heart disease risk with high clinical reliability.

📊 Task Report & Findings

1. Exploratory Data Analysis (EDA)

• Data Inspection: Utilized .head(), .info(), and .describe() to verify 303 patient records and 14 clinical features.
• Visualizations: - Histograms: Revealed a peak in heart disease cases for patients aged 50–60.
  • Boxplots: Identified significant outliers in Cholesterol (chol) and Resting Blood Pressure (trestbps).
• Handling Outliers: Implemented StandardScaler to normalize feature ranges, ensuring extreme values (like high cholesterol) didn't skew the model's coefficients.

2. Machine Learning Model

• Algorithm: Logistic Regression (Classification).
• Setup: 80% Training Data / 20% Testing Data.
• Optimization: Resolved convergence warnings by increasing max_iter and applying feature scaling for mathematical efficiency.

3. Evaluation Results

The model achieved high performance, proving its reliability for medical screening:

• Final Accuracy: 85.25%
• F1-Score: 0.86 (indicates a strong balance between Precision and Recall).
• Confusion Matrix Performance:
  • Correctly identified 25/29 healthy cases.
  • Correctly identified 27/32 disease cases.

🛠️ Setup Instructions

To run this project on your local machine:

1. Clone the repository.
2. Activate your virtual environment: source venv/bin/activate
3. Install dependencies: pip install -r requirements.txt
4. Run the notebook: jupyter notebook Model_Evaluation.ipynb

📧 Contact

• Email: syedmuhammadshoaibamjad@gmail.com