Embedded detection of atrial fibrillation and atrial flutter in single-channel ECG
| Main Author: | Auer, Eric |
|---|---|
| Format: | Report Journal |
| Bahasa: | eng |
| Terbitan: |
, 2020
|
| Subjects: | |
| Online Access: |
https://zenodo.org/record/4311642 |
Daftar Isi:
- This project report describes a proof of concept stand-alone embedded system, based on an ESP32 WROOM32 controller module and the Maxim MAX30003 ECG AFE, which records and immediately processes single-channel ECG to determine whether the user is likely to be experiencing an episode of atrial fibrillation or atrial flutter at the time of the measurement. The algorithm devised for this task improves the signal to noise ratio by selecting a number of heart beats from the recording to compute an improved average ECG. From this ECG, morphology metrics are extracted and rated. In addition, the system displays some R/R metrics and uses a periodogram score, based on the derivative of the signal and weighted by frequency bands, as supplemental information source for the classification task. The software written for this project will be made available under the open source FreeBSD license: It should be easy to port to existing hardware platforms such as the HeartyPatch and the custom ecghelper embedded system of the author, which will be released as open hardware later.
- Project Report, Neural Engineering, University of Applied Sciences Saarbrücken (htwsaar) Supervisor: Prof. Dr. Oliver Scholz