Towards understanding ECG rhythm classification using convolutional neural networks and attention mappings
Access to electronic health record (EHR) data has motivated computational advances in medical research. However, various concerns, particularly over privacy, can limit access to and collaborative use of EHR data. Sharing synthetic EHR data could mitigate risk. In this paper, we propose a new approach, medical Generative Adversarial Network (medGAN), to generate realistic synthetic patient records. Based on input real patient records, medGAN can generate high-dimensional discrete variables (e.g., binary and count features) via a combination of an autoencoder and generative adversarial networks. We also propose minibatch averaging to efficiently avoid mode collapse, and increase the learning efficiency with batch normalization and shortcut connections. To demonstrate feasibility, we showed that medGAN generates synthetic patient records that achieve comparable performance to real data on many experiments including distribution statistics, predictive modeling tasks and a medical expert review. We also empirically observe a limited privacy risk in both identity and attribute disclosure using medGAN.
PDF AbstractCode
Datasets
Results from the Paper
Ranked #1 on
Arrhythmia Detection
on The PhysioNet Computing in Cardiology Challenge 2017
(Accuracy (TRAIN-DB) metric)
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Arrhythmia Detection | The PhysioNet Computing in Cardiology Challenge 2017 | Towards Understanding ECG Rhyth | Accuracy (TRAIN-DB) | 88% | # 1 |
