EF-estimation-ECG

EF-estimation-ECG is a Python-based project aimed at estimating left ventricular ejection fraction (EF) from 12-lead electrocardiogram (ECG) data.

Features

• Data Processing Utilities: Functions for loading, preprocessing, and managing ECG datasets (data_utils.py).
• Model Architectures: Implementation of deep learning models tailored for ECG data analysis, including ResNet-based models and MLP (models.py) and transformer-based models (ecgformer.py).
• Training and Evaluation Pipelines: Scripts to train models (main.py, main_sklearn.py) and evaluate their performance (eval.py).
• Dataset Generation: Tools to format datasets for training and testing (generate_dataset.py).
• Interpretability and Visualization: Modules to interpret model predictions and visualize ECG data (interpretability/, visualizations/).

Installation

To set up the project environment, follow these steps:

1. Clone the Repository:

   git clone https://github.com/alinadevkota/EF-estimation-ECG.git
   cd EF-estimation-ECG

2. Create a Virtual Environment (optional but recommended):

   python3 -m venv venv
   source venv/bin/activate

3. Install Required Dependencies:

   pip install -r requirements.txt

Usage

Data Preparation

Ensure your ECG dataset is organized appropriately. The data_utils.py module provides functions to assist with data loading and preprocessing. Modify paths and parameters as needed to align with your data storage.

Training a Model

To train a model using the default settings:

python main.py

Evaluating a Model

After training, evaluate the model's performance:

python eval.py --model_path path/to/saved_model.pth --data_path path/to/test_data/