- Published on
In this article, I explore a novel deep learning architecture from a recent publication of mine that integrates the H∞ filter from control theory into a CNN-LSTM framework to enhance robustness against noise and class imbalance in arrhythmia detection from heart sound recordings. Our approach achieves state-of-the-art performance on the PhysioNet CinC Challenge 2016 dataset, demonstrating its potential for scalable and reliable cardiac screening.
