Skip to content

ShayanRiyaz/Heart_Rhythm_Analysis

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

44 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Heart Rhythm Analysis

By: Shayan Riyaz

Motivation:

The purpose of this project is to explore Heart Rhythms using:

  1. Signal Processing Techniques
  2. Machine/Deep Learning
  3. Causal Analysis/Inference

Dataset

Currently I am using ~450 subject MIMIC III subject via physionet. I converted Peter Charlton's collate_mimic_perform_af_dataset script for extracting the data from Matlab to python. (Using a mix of LLMs and my own intuiton). Next I used the MSPTDfast peak detection algorithm (again by Peter Charlton & again converted using LLM and my own understanding of the algorithm.).

Datasets Prepared:

Model Development

  • Currently i created a dummy Conv1D net for peak detection for PPG Signals. (not sure if its necessary but a good learnign exercise).

Result Version 1 - Conv1d Net

Current Status

Result Version 2 - Denoiser and Peak Extractor

Goals:

  • Noisy Frame Detection using Deep Learning + Simple Statistical Measurments
  • Peak Detection using Semi-supervised (pre-labeled using MSPTDfast algorithm) + Reinforcement learning based on physiological behaviour of pleth.
  • Calculation of Features such as:
    • Time-domain: HRV (Heart Rate Variability), mean HR, PTT (Pulse Transit Time)
    • Morphological: Pulse amplitude, rising time, dicrotic notch position
    • ? (Still Studying)
  • Some Type of Causal Modeling Architecture (X features, treatment) → outcome using TarNet
  • Purpose of Causal Modeling: Predicting patient deterioration, alarm risk, personalization

References

Dataset Reference:

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages