New tool enables early intervention for diabetes related foot ulcers

A new thermal-imaging AI tool called THERMUL has been developed by a collaboration between the University of Melbourne and RMIT University to help prevent hospitalizations and amputations related to diabetes-related foot ulcers. Diabetes affects over 500 million people globally, with up to 34% developing a foot ulcer during their lifetime.

These ulcers can lead to infection, hospitalization, amputation, and early death if not managed properly. Using thermal imaging and AI, THERMUL can predict whether a foot ulcer is likely to heal within 12 weeks by analyzing patterns within thermal images that indicate healing progress or risk of delayed healing. Researchers conducted a clinical study involving 110 patients at Austin Health and St Vincent's Hospital Melbourne, taking images of their ulcers at initial presentation and follow-up visits.

The thermal image data was used to train an AI algorithm to recognize heat patterns associated with ulcers that healed versus those that did not. Endocrinologist Professor Elif Ekinci and RMIT professor Dinesh Kumar believe that THERMUL has the potential to improve health equity and reduce the estimated $875 million annual cost of diabetes-related foot disease to the Australian healthcare system by enabling earlier identification of ulcers at risk of delayed healing. However, further validation is required before the tool can be used clinically.