ECG Signal Classification Project

Overview

This project aims to classify ECG signals as either Normal or Left Bundle Branch Block (LBBB) using machine learning models.
The key steps involved include:

• Data preprocessing
• Feature extraction
• Model training and evaluation
• Deployment

---

Introduction

This project leverages machine learning techniques to classify ECG signals as Normal or LBBB. It involves the following steps:

• Preprocessing ECG signals using Butterworth bandpass filters.
• Extracting features from the signals using wavelet transforms.
• Training and evaluating different machine learning models.
• Deploying the best model using a GUI application.

---

Data Preprocessing

The data preprocessing step involves:

1. Removing noise from the ECG signals using a Butterworth bandpass filter.
2. Normalizing the signals to a standard range to ensure consistent feature extraction.

---

Feature Extraction

Features are extracted from the preprocessed signals using wavelet transforms.
Statistical features such as:

• Mean
• Standard deviation
• Skewness
• Kurtosis
  are calculated from the wavelet coefficients.

---

Model Training and Evaluation

Various machine learning models are trained and evaluated on the extracted features, including:

• K-Nearest Neighbors (KNN)
• Support Vector Machine (SVM)
• Random Forest

The best model is selected based on accuracy and other evaluation metrics.

---

Deployment

The best-performing model (KNN) is deployed using a GUI application.
This application allows users to:

• Input ECG signals
• Receive a classification result: Normal or LBBB

---

Usage

To use the deployed model:

1. Run the gui.py file to start the GUI application.
2. Input your ECG signal into the application.
3. View the classification result (Normal or LBBB).

---

Results

The project achieved impressive results:

• The best model (KNN) achieved an accuracy of 98.75% on the test set.
• Detailed evaluations and feature importances are documented in the project report.