MM466 Project: ECG Signal Classification for Arrhythmia Detection using Machine Learning in MATLAB

MM466 Project | ECG Signal Classification for Arrhythmia Detection using Machine Learning in MATLAB

In this MM466 project, we explore how machine learning can be applied to biomedical signals—specifically, ECG data—to detect and classify cardiac arrhythmias. Using MATLAB, we developed a complete signal processing and classification pipeline that includes data cleaning, feature extraction, dimensionality reduction (PCA), class balancing using SMOTE, and model training with classifiers like Random Forest, k-NN, Decision Tree, and Multilayer Perceptron (MLP).

Key Highlights:
✅ MIT-BIH Arrhythmia Database preprocessing
✅ R-peak detection and feature extraction
✅ SMOTE for handling class imbalance
✅ Neural Network training and evaluation
✅ Confusion matrices, ROC curves, and model comparison
✅ Integration with the Classification Learner App
✅ Discussion on how this project connects with Mechanical Engineering

Final model performance: Random Forest achieved 98.53% accuracy, with SMOTE significantly improving minority class detection.

Whether you're a student, engineer, or someone interested in biomedical signal processing and AI, this video is for you guys.

This video provides a overview of the project and shows how to run the codes.
🔔 Don’t forget to like, subscribe, and share!

📁 Project Code & Report:
GitHub: [https://github.com/Mrun-mayi/ecg-ml-blog]
Medium Blog: [https://medium.com/@s11208621/mm466-blog-8f43aa6381ba​]

#ECGClassification #MATLAB #MachineLearning #ArrhythmiaDetection #SMOTE #MechanicalEngineering #MITBIH #ECGSignalProcessing #NeuralNetwork #BiomedicalAI

Видео MM466 Project: ECG Signal Classification for Arrhythmia Detection using Machine Learning in MATLAB канала Mrunmayi Badgujar