Skip to content

Rlign v1.0 Release
Compare
Choose a tag to compare
@l-bick l-bick released this 18 Jul 14:59
· 3 commits to main since this release
v1.0
3fe63ee
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Learn about vigilant mode.

We are excited to announce our new approach to ECG signal processing, Rlign, leveraging shallow learning techniques. Our method focuses on enhancing explainability, reducing large training data requirements, and simplifying adaptation for unsupervised tasks. The highlights include:

  • Transformation Process: Restructures ECG signals into a fully structured format.
  • Alignment & Resampling: Aligns R-peaks across all signals and resamples segments between R-peaks, with both heart rate-dependent and independent options.
  • Enhanced Traditional Analysis: Facilitates classification, clustering, and explainability, outperforming commercial software and CNN approaches.
  • Limited Data: Particularly effective when dealing with limited training datasets.
  • Code Framework: A fully tested and publicly accessible code framework for robust alignment pipelines.
Assets 2
Loading