A state-of-the-art review on computational methods for predicting the occurrence of cardiac autonomic neuropathy

Cardiac autonomic neuropathy (CAN) is a serious complication of type 2 diabetes mellitus (T2DM) that is associated with an increased risk of cardiovascular mortality. Nonetheless, it is a common and often underdiagnosed complication of T2DM. CAN occurs due to damage to the autonomic nervous system and manifests as coronary vessels ischemia, arrhythmias, silent myocardial infarction, severe orthostatic hypotension, and sudden death syndrome. Clinical approaches for evaluating CAN include assessment of symptoms and signs, cardiovascular reflex tests based on heart rate and blood pressure known as Ewing's test, electrocardiography, and heart rate variability. In recent years, there has been an extraordinary advancement in AI-assisted diagnostics. AI-assisted diagnosis and prediction of CAN based on Ewing's test and other clinical features could facilitate the early detection of CAN. Further research utilizing large-scale databases, novel algorithms and features could further enhance the performance of AI-assisted diagnosis and prediction of CAN.