This repository contains the source code and associated files for the research study titled "Learning with Incomplete Labels of Multisource Datasets for ECG Classification".
This research study focuses on addressing the challenge of classifying electrocardiogram (ECG) data from multiple sources with incomplete labels. The study proposes a deep-learning-based framework to improve the accuracy of ECG classification in such scenarios. This repository contains the source code and resources related to the research.
These instructions will help you get a copy of the project up and running on your local machine for development and testing purposes.
Before you begin, ensure you have met the following requirements:
- Python (>=3.6)
- TensorFlow (>=2.0)
- Other dependencies (listed in
requirements.txt
)
-
Clone the repository:
git clone https://github.com/sdnjly/multisource-ecg-classification.git cd multisource-ecg-classification
-
Create a virtual environment (recommended):
python -m venv venv source venv/bin/activate
-
Install required packages:
pip install -r requirements.txt
The datasets used in this study can be found from the Physionet/CinC challenge 2020/2021 websites. You can follow instructions on the websites to downloads the datasets.
-
model training
python train_model.py training_data model
training_data
is a folder of training data files,model
is a folder for saving your models,test_data
is a folder of test data files -
model testing
python test_model.py model test_data test_outputs