Skip to content

Latest commit

 

History

History
25 lines (15 loc) · 1.33 KB

README.md

File metadata and controls

25 lines (15 loc) · 1.33 KB

12 Leads ECG Signal Reconstruction from Single Lead

12 Leads ECG is essential for examining cardio status. However, most commercial-grade products are unable to capture full 12 Leads ECG signal. And so, Reconstructed 12 Leads ECG Signal might be a better solution for identifying arrhythmia for data collected by smart devices. Here I bulid a Deep Learning model with U-Net Architecture for this specific task.

Here are reconstructed ECG charts of this model. ID_140 ID_547

Chart of Losses on each epochs U-Net_Loss_Result

Dataset Description

I built this model with Code-15% dataset. You can find the description and original data from following link. https://zenodo.org/record/4916206

You can also download cleaned data for this particular project via following Link: https://1drv.ms/u/s!ArQCikHAsFj6oPJoQY7h9rIggjoaug?e=YjNprR

Reference:

  1. Learning to See in the Dark: https://arxiv.org/abs/1805.01934

  2. I am sorry that I lost the original source of my U-Net Implementation. Please let me know if you are the author.