Prediction of Diabetic Neuropathy Using Machine Learning Techniques

Abstract

Peripheral polyneuropathy is one of the most common complications in patients with diabetes mellitus, and it results in neuropathic pain, falling tendency, and foot ulcers as well as sensory and motor impairments. Numerous risk factors for diabetic neuropathy had been revealed through statistical analysis; however, statistics draw population inferences and might not be suitable for providing realtime prediction for each patient in clinical practice. Machine learning techniques were developed to find any predictive patterns based on input data. Such strategies can help predict neuropathy in diabetic patients, enabling prevention or early treatment to increase quality of life in diabetic patients. This article summarizes recent studies concerning the prediction of diabetic neuropathy using machine learning techniques, and suggests approaches for useful translation of these methods in the medical field.