AI and Angiosome Based Analysis of Diabetic Foot Thermal Images for the Diagnosis of Ulcer Risk

Infrared thermography plays a significant role in the medical field due to its wide application in diagnosing various diseases, including the detection of diabetic foot issues. This paper presents a novel method for classifying diabetic foot patients into ulceration risk groups based on contralateral variations in plantar angiosome temperatures. A total of 140 type II diabetic patients without ulceration or amputation were included in this study and classified into three ulcer risk groups. Foot angiosome temperatures were measured using an AI mobile application developed for this study, which employs the DE-ResUnet architecture for image segmentation and the Iterative Closest Point method for image registration. Our results indicate that this method effectively distinguishes between moderate-risk and high-risk patients for ulcer development with a significance level of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$p$</tex> < 0.01. Infrared thermography, combined with the advanced capabilities of our AI mobile application, proves to be a useful tool for the early diagnosis and prevention of diabetic foot complications.