Resources: ECG processing and features

[DominiqueMakowski]

Collection of resources to help us getting the ECG aspect right, feel free to add more:

---

R-peak detection & Segmentation

• See #77

Artifacts

• [x] Lipponen & Tarvainen, 2019

  HRV

General

• Guidelines for Reporting Articles on Psychiatry and Heart rate variability
• Slides by @michagaebler (2020) on HRV

Complexity

• Entropy analysis of heart rate variability and its application to recognize major depressive disorder
• Entropy Analysis of Short-Term Heartbeat Interval Time

Other

• [ ] ECG-RSP phase-amplitude coupling

[JanCBrammer]

To get us started with ECG processing I'm happy to add an ECG peak detection algorithm to the ecg module. I recently benchmarked the algorithm on the Glasgow University Database (GUDB) (see 5. Changelog for details). Then the users could already get heart rate and -period out of their ECG data and we would have a basis for building HRV analyses. If this sounds like a good idea I'll get working on a PR soonish in the new year :)

On a related note, here's a nice overview of popular ECG peak detection algorithms and their performance on established benchmarks.

[DominiqueMakowski]

@JohnDoenut That's great 😍 Looking forward to your PR in the next decade 😁

[DominiqueMakowski]

@Tam-Pham It would be useful to get an overview of the existing pipelines for ECG cleaning (parameters of filters etc..). For instance, biosspy uses only simple filtering:

https://github.com/PIA-Group/BioSPPy/blob/e65da30f6379852ecb98f8e2e0c9b4b5175416c3/biosppy/signals/ecg.py#L69-L76

But there are more packages and surely some publications about it :wink:

[JanCBrammer]

See #65.

[DominiqueMakowski]

Closing this in favour of the more specialized issues.