Quantum inspired improved AI computing for the sensors of cardiac mechano-biology

Abstract

Recent advancements in the field of quantum mechanics have opened the research avenues for the accurate mapping of the heart’s conductivity with the aid of quantum sensors.The heart rate variability can now be accessed to better levels with the aid of recent noninvasive approaches. This novel idea can be improved with the aid of transfer learning where a smart machine learning algorithm, the “improved dynamic time wrapping” can be used, after selecting the important attributes such as the cardiac action potential. During this research, a two step approach is used for the dynamical analysis of the atrial fibrillation cardiac amyloid, since the Atrial-fibrillation (AF) is more common in patients with the “cardiac amyloidosis”, resulting from the deposition of amyloid protein. During the first step, inspired from the experimental studies in the domain of cardiac amyloidosis, an electric heart model is simulated for different amyloid deposition levels, and during the second step, the transfer learning algorithm is applied to explore the time series data of the electric potential and its divergence. This AI- synchronised divergence can be used to cross verify the electric potential divergence, detected by novel quantum sensors, in ongoing and future research projects.