Heart Disease Prediction using Machine Learning

1. Introduction

• The project aims to predict heart disease using machine learning models based on medical data.
• The dataset includes patient details such as age, cholesterol levels, blood pressure, and ECG results.

---

2. Data Preprocessing

• Data Cleaning: Checked for missing values and handled any inconsistencies.
• Feature Scaling: Standardized numerical data to improve model performance.
• Data Splitting: Divided the dataset into training and testing sets (80-20 split).

---

3. Handling Imbalanced Data

• SMOTE (Synthetic Minority Oversampling Technique) was used to balance the dataset.
• Feature Selection and Dimensionality Reduction (PCA, t-SNE) were applied to enhance model efficiency.

---

4. Model Selection & Training

• Evaluated multiple machine learning models:
  • Logistic Regression
  • Random Forest
  • Support Vector Machine (SVM)
  • K-Nearest Neighbors (KNN)
  • Neural Networks
• Used performance metrics like Accuracy, Precision, Recall, and AUC-ROC.

---

5. Conclusion

• Random Forest with SMOTE performed best, achieving around 96.8% accuracy.
• Feature Engineering & PCA improved model performance.
• The project demonstrates how machine learning can assist in early detection of heart disease, potentially helping doctors make better decisions.

---