https://github.com/MousaviSajad/ECG-Heartbeat-Classification-seq2seq-model

Inter- and intra- patient ECG heartbeat classification for arrhythmia detection: a sequence to sequence deep learning approach
https://github.com/MousaviSajad/ECG-Heartbeat-Classification-seq2seq-model

biosignals cnn deep-learning ecg ecg-heartbeat-classification sequence-to-sequence tensorflow

Last synced: about 1 month ago
JSON representation

Inter- and intra- patient ECG heartbeat classification for arrhythmia detection: a sequence to sequence deep learning approach

• Host: GitHub
• URL: https://github.com/MousaviSajad/ECG-Heartbeat-Classification-seq2seq-model
• Owner: MousaviSajad
• License: other
• Created: 2019-02-11T02:18:24.000Z (over 5 years ago)
• Default Branch: master
• Last Pushed: 2023-06-13T23:18:13.000Z (about 1 year ago)
• Last Synced: 2024-02-23T16:12:40.025Z (5 months ago)
• Topics: biosignals, cnn, deep-learning, ecg, ecg-heartbeat-classification, sequence-to-sequence, tensorflow
• Language: Python
• Homepage:
• Size: 706 KB
• Stars: 162
• Watchers: 6
• Forks: 56
• Open Issues: 10
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Lists

• awesome-ai-cardiology - SajadMo/ECG-Heartbeat-Classification-seq2seq-model - Inter- and intra- patient ECG heartbeat classification for arrhythmia detection: a sequence to sequence deep learning approach (Code / Repositories)

README

        
# Inter- and intra- patient ECG heartbeat classification for arrhythmia detection: a sequence to sequence deep learning approach

# Paper

 Our paper can be downloaded from the [arxiv website](https://arxiv.org/pdf/1812.07421v2)

 * The Network architecture

  ![Alt text](/images/seq2seq_b.jpg)

 

## Requirements

* Python 2.7

* tensorflow/tensorflow-gpu

* numpy

* scipy

* scikit-learn

* matplotlib

* imbalanced-learn (0.4.3)

## Dataset

We evaluated our model using [the PhysioNet MIT-BIH Arrhythmia database](https://www.physionet.org/physiobank/database/mitdb/)

* To download our pre-processed datasets use [this link](https://drive.google.com/drive/folders/19bDrAqlSGQuNLRmA-7pQRU9R81gSuY70?usp=sharing), then put them into the "data" folder.

* Or you can follow the instructions of the readme file in the "data preprocessing_Matlab" folder to download the MIT-BIH database and perform data pre-processing. Then, put the pre-processed datasets into the "data" folder.

## Train

* Modify args settings in seq_seq_annot_aami.py for the intra-patient ECG heartbeat classification

* Modify args settings in seq_seq_annot_DS1DS2.py for the inter-patient ECG heartbeat classification

* Run each file to reproduce the model described in the paper, use:

```

python seq_seq_annot_aami.py --data_dir data/s2s_mitbih_aami --epochs 500

```

```

python seq_seq_annot_DS1DS2.py --data_dir data/s2s_mitbih_aami_DS1DS2 --epochs 500

```

## Results

  ![Alt text](/images/results.jpg)

## Citation

If you find it useful, please cite our paper as follows:

```

@article{mousavi2018inter,

  title={Inter-and intra-patient ECG heartbeat classification for arrhythmia detection: a sequence to sequence deep learning approach},

  author={Mousavi, Sajad and Afghah, Fatemeh},

  journal={arXiv preprint arXiv:1812.07421},

  year={2018}

}

```

## References

 [deepschool.io](https://github.com/sachinruk/deepschool.io/blob/master/DL-Keras_Tensorflow)

 

## Licence 

For academtic and non-commercial usage