Thermal Imaging and Deep Learning For Neuropathy Risk Detection in Diabetic Feet

Diabetic foot complications pose significant health risks, leading to infections and amputations. Early detection is crucial for effective management. This study utilizes thermal imaging technology to analyze temperature differences between feet as an early indicator of complications. Focusing on three parameters—overall thermal pattern, average temperature, and inter-foot temperature difference—we identified significant distinctions among three groups: Control, Diabetic Non-Neuropathic, and Diabetic Neuropathic patients. A novel plantar thermogram database was created from thermal images collected in Annaba, Algeria, contributing valuable region-specific data. Additionally, a user-friendly web application was developed to facilitate the analysis of thermal images, improving accessibility for healthcare professionals and enhancing patient care.