This repository contains a machine learning project focused on the application of Convolutional Neural Networks (CNNs) to synthesize Electrocardiogram (ECG) signals from the PTB-XL dataset. The project employs Python and the PyTorch library for the design, training, and testing of initial CNN models.
The main goal of this project is to explore the capabilities of CNNs in generating realistic ECG signals, in the case where if a lead is blocked in clinical practice, medical devices could potentailly generate missing leads. It also paves the way in developing latent spaces which understand how the 12 ECG signals relate to one another, and thus assisting in the hard-to-diagnose diseases such as left-branch-bundle-block. The project involves setting up a robust testing environment and implementing a CNN architecture inspired by recent research in the field.
Keywords: Convolutional Neural Networks, CNN, ECG Synthesis, Python, PyTorch, Machine Learning, Deep Learning, ML Ops, ML Engineering, Testing Environment, Model Design, GitHub