Skip to content

Multi-module Recurrent Convolutional Neural Network with Transformer Encoder for ECG Arrhythmia Classification

Notifications You must be signed in to change notification settings

Vidhiwar/multimodule-ecg-classification

Repository files navigation

multimodule-ecg-classification

RESEARCH-PAPER:{

  TITLE: "Multi-module Recurrent Convolutional Neural Network with Transformer Encoder for ECG Arrhythmia Classification",

  CITE:  "https://ieeexplore.ieee.org/abstract/document/9508527",
  
  YEAR: 2021,
  
  CONFERENCE: "IEEE EMBS",
  
  AUTHORS: ["Duc Le^",
                "Vidhiwar Singh Rathour^",
                      "Sang Truong", 
                               "Quan Mai^, 
                                        Patel Brijesh; 
                                                Ngan Le"],
                                                      ^: Equal Contribution}  

DIRECTORY-TREE:{

 data: "Directory: Datasets for training are stored here.",

 utils: "Directory: Utility based files",
 
 examples: {"Directory: github/awni/ecg/":[
        "Cardiologist-level arrhythmia detection and classification in ambulatory electrocardiograms using a deep neural network"]},

 models:{ "Directory: DNN Models": {
        resnet_cnn.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, CNN, Word2Vec",
        resnet_lstm_phy2017.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN",
        resnet_lstm.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN",
        resnet_w2v.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN, Word2Vec",
        resnet_lstm_mitbih.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN, Word2Vec, Attention AYN"}},

 ecg_cnn.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, CNN, Resnet",
 ecg_w2v.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, Word2Vec, Resnet",
 ecg_lstm.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN, LSTM",
 ecg_phy2017.py:"Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN",
 ecg_mitbih.py: "Python: Pytorch implememntation of Multimodal-ecg-classification, LSTM, CNN, Word2Vec, Attention AYN",
 transform_data.ipynb: "Jupyter Notebook: Python implementation for Data Generation, and Preprocessing"}

HOW-TO-USE:{

  Uno: "Make sure the required libraries (Torch, Panda, Tqdm, ... etc.,). are installed",
  Dos: : "Use the examples directory to download and preprocess data.",
  Tres: "Follow transform_data.ipyn to get data ready for training.",
  Cuatro: "Run python ecg_###.py to train on training data, and validate on validation data",
  Cinco: "By default results are saved in checkpoints directory"}

IMAGES:{

Model.png Results.png}

EOF

About

Multi-module Recurrent Convolutional Neural Network with Transformer Encoder for ECG Arrhythmia Classification

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published