Advertisement

RMIT University thermal imaging offers early help for diabetic foot ulcers

AI-assisted imaging predicts if diabetes-related ulcers are likely to heal.

19 November 2026


The AI tool uses thermal images to predict likelihood of healing. Credit: RMIT University.Surveys show that about one-third of diabetes sufferers will develop a diabetic foot ulcer (DFU) during their life, leading to possible hospitalization or even lower limb amputation.

Early detection and diagnosis of these ulcers is a key clinical issue, and optical technologies have played their part.

Examples include a device based around structured light developed by California’s Modulim, in which spatial frequency domain imaging (SFDI) measures tissue oxygenation and perfusion at the point-of-care; and research by Israel’s IR-MED into whether infra-red reflectance spectroscopy can detect the differences in absorption and reflection of diseased tissues compared to healthy tissues.

A project at Australia’s RMIT University has now demonstrated a new smartphone-based AI tool called TherMUL, intended to provide clinicians with an indication of healing risk much earlier by analyzing thermal images captured when a patient first presents for care.

“A fundamental limitation of current best practice is its reliance on retrospective assessment of healing progression,” commented the project in its Journal of Diabetes Science and Technology paper. “There is, therefore, an urgent need for objective, clinically feasible methods to identify DFUs at risk of delayed healing at the first clinical encounter.”

Thermal imaging is an established method of analyzing the condition of the skin surface and the tissues immediately underneath. But most applications of thermal imaging to DFUs have relied on lateral temperature comparison, with temperature asymmetry between the affected foot and the other, presumably unaffected, foot interpreted as a marker of pathology.

RMIT envisaged a solution that did not need external references, but instead used current image texture analysis techniques to extract higher-order spatial information from thermal images, quantifying clinically important heterogeneity in the data not apparent from mean temperature alone.

Advertisement

Improved access to timely care

RMIT’s TherMUL prototype is a smartphone-based thermal imaging platform incorporating a commercially available FLIR thermal camera and feeding the data into a custom image analysis pipeline for automated wound assessment. The group's AI model was trained to identify heat patterns associated with ulcers that healed successfully and those that did not.

“These heat patterns may provide an early indication of a wound’s likely healing trajectory, potentially allowing clinicians to identify people who might benefit from earlier intervention or escalation of care,” said Dinesh Kumar from RMIT’s School of Engineering.

In trials researchers took images of patients’ ulcers when they initially presented to the clinic, with additional images captured two and four weeks later. At 12 weeks, clinicians assessed whether each ulcer had healed, allowing the research team to compare healing outcomes with the thermal image data collected earlier.

TherMUL achieved a 75 percent sensitivity for predicting unhealed DFUs, and a specificity of 53 percent for identifying healed DFUs, commented RMIT. In the preliminary evaluation, 5 of 7 predictions from the TherMUL evaluation correctly matched the 12-week podiatry outcome.

Further validation in larger and more diverse real-world settings is required to prove TherMUL’s clinical potential, particularly given known challenges around thermography standardization and environmental sensitivity. But the initial findings show the value of accessible, point-of-care decision support in the routine management of DFUs, according to RMIT.

“Diabetes-related foot disease places a significant burden on individuals and the Australian health system, costing an estimated $875 million every year,” said Elif Ekinci from the University of Melbourne. “By enabling earlier identification of ulcers at risk of delayed wound healing, TherMUL has the potential to improve access to timely care and reduce complications.”

Advertisement
Latest Stories
Article Tags
Advertisement
Advertisement