Physiowise: A Physics-Aware Approach to Dicrotic Notch Identification

Dicrotic Notch (DN), one of the most significant and indicative features arterial blood pressure (ABP) waveform, becomes less pronounced thus harder to identify as a matter aging pathological vascular stiffness. Generalizable automatic DN identification for such edge cases is even more challenging in presence unexpected ABP waveform deformations that happen due internal external noise sources or conditions cause hemodynamic instability. We propose physics-aware approach, named Physiowise (PW), first employs cardiovascular model augment original reduce deformations, then apply set predefined rules on augmented signal find locations. have tested proposed method in-vivo data gathered from 14 pigs under hemorrhage sepsis study. Our result indicates \(52\% \) overall mean error improvement with \(16\% higher detection accuracy within lowest permitted range 30 ms . An additional hybrid methodology also allow combining augmentation any application-specific user-defined rule set.