Deep Learning R-Wave Detection for Electrocardiographic Gating in Cardiac MRI

This repository contains code for developing an R-wave detection convolutional neural network.

Published article: https://doi.org/10.1148/ryct.250104

File Descriptions

• README.md: This file, providing an overview of the project and its components.
• requirements.yml: Conda environment configuration file listing all dependencies.
• ecg_data_processor.py: Processes raw ECG data from different sources and saves the processed data for further analysis.
• ecg_dataset_manager.py: Manages various ECG datasets, including loading signals and annotations, and preparing data for model training.
• ucsd_ecg_dataset.py: Manages the UCSD ECG dataset, including loading and processing trigger data files.
• process_ecg_waveforms.py: Format raw ECG waveform data and saves the processed data using ecg_data_processor.py.
• process_ecg_annotations.py: Prepares the training dataset from UCSD data and Label Studio annotations.
• train_detector.py: Script to train the deep learning model for QRS detection.
• deep_qrs_detector.py: Defines a deep learning model for detecting QRS complexes in ECG signals and includes methods for training and evaluating the model.
• qrs_detection_timer.py: Measures the execution time of different QRS detection methods.
• labelstudio_tools.py: Tools for handling Label Studio annotations and converting them to dataframes.
• beat_classification_functions.py: Functions for classifying ECG beats based on detected R-peaks and RR-intervals.
• beat_classifier.py: Script to classify ECG beats algorithmically and save the results.