❤️ ECG Heartbeat Classification (Machine Learning & Deep Learning)

An end-to-end project that classifies ECG signals into normal and abnormal heartbeats using both Machine Learning and Deep Learning algorithms.

---

🧠 Project Overview

This project demonstrates how machine learning and deep learning can be used to detect cardiac abnormalities from ECG signals. It compares multiple models to evaluate accuracy, recall, and F1-score, aiming to assist in early detection of arrhythmia.

---

⚙️ Tech Stack

• Languages: Python
• Libraries: Pandas, NumPy, Scikit-learn, TensorFlow, Keras, Matplotlib, Seaborn
• Environment: Jupyter Notebook / VS Code

---

🧾 Dataset

• Source: MIT-BIH Arrhythmia Database
• Features: Heartbeat signal data (time series)
• Target: Heartbeat class (Normal, PVC, LBBB, RBBB, etc.)

---

🧩 Methodology

1. Data Cleaning & Normalization
2. Signal Segmentation & Feature Extraction
3. Model Building:
   • Logistic Regression
   • K-Nearest Neighbors (KNN)
   • Random Forest
   • Convolutional Neural Network (CNN)
   • Long Short-Term Memory (LSTM)
4. Model Evaluation using Accuracy, F1-score, ROC-AUC
5. Comparison between ML & DL models

---

📈 Results

Model	Accuracy	F1-Score	ROC-AUC
Random Forest	90%	0.88	0.91
CNN	95%	0.93	0.96
LSTM	97%	0.95	0.98

✅ Best Model: LSTM (Deep Learning)

---

📂 Files Included

File	Description
ecg_heartbeat_classification.ipynb	Full code with EDA, ML, and DL models
ecg_heartbeat_data.csv	Dataset used for model training
ECG_Classification_Report.docx	Detailed project report
ECG_Insights.pdf	Summary report with charts and visual insights

📘 All results and visual outputs are available in the report files.

---

🚀 Future Enhancements

• Integrate real-time ECG signal input
• Develop a web-based dashboard using Streamlit or Flask
• Deploy model for live arrhythmia detection

---

👨‍💻 Author

Hemant Kumar M
📍 Newcastle upon Tyne, UK
📧 hihemantkumar786@gmail.com
🔗 GitHub: github.com/Hemant-Kumar786