Whole-foot plantar temperature distributions can descriptively characterize differences among healthy individuals, patients with isolated DM, and those with combined PAD and DM, with healthy controls showing symmetric thermal footprints, DM showing globally elevated compact profiles, and combined PAD_DM showing mixed asymmetric patterns.
Healthy controls predominantly showed symmetric thermal footprints, with the majority assigned to healthy or borderline healthy categories.
The DM cohort frequently exhibited globally elevated and compact temperature-distribution profiles with heterogeneous category assignments.
The combined PAD_DM cohort showed the highest agreement with its assigned category, with three-quarters of participants correctly classified.
The isolated PAD cohort was the most thermally heterogeneous group, with none assigned to an isolated PAD output category.
The study developed a cohort-derived deterministic rule-based model using bilateral kernel density estimation profiles to characterize plantar thermal patterns.
PAD and DM may produce different and overlapping plantar thermal patterns that can be descriptively differentiated using infrared thermography.
This research suggests that infrared thermography — a non-invasive imaging technique that measures heat distribution on the skin surface — can detect distinct temperature patterns on the soles of the feet in people with different vascular and metabolic conditions. The study examined 73 participants divided into four groups: healthy individuals, those with peripheral arterial disease (PAD, a condition of reduced blood flow to the limbs), those with diabetes mellitus (DM), and those with both conditions combined. By analyzing full-foot temperature maps, researchers found that healthy feet tend to show symmetric (mirror-image) temperature patterns between left and right feet, while diabetic feet tend to show uniformly elevated temperatures across the whole foot, and feet affected by both PAD and diabetes show irregular, asymmetric mixed patterns. A notable finding was that the isolated PAD group was the most difficult to characterize — none of these participants produced a thermal profile matching a distinct 'PAD-only' pattern, and nearly half were classified as resembling the combined PAD and diabetes group. This suggests that PAD alone may produce highly variable thermal signatures, possibly making it harder to detect with this method in isolation. The researchers developed a rule-based model to systematically assign thermal patterns to categories, though they emphasized this model was built from the same data used to test it and should not be considered a validated diagnostic tool. This research suggests that infrared thermography of the feet could eventually assist clinicians in non-invasively screening for or monitoring conditions like diabetes and peripheral arterial disease, particularly when imaging the full foot rather than specific regions. However, since this was a small, exploratory, single-center study without an independent validation set, further research with larger and more diverse populations would be needed before this approach could be considered for clinical use.
Tanasković S, Milačić A, Ružić N, Graovac S, Petrović J, Bojić M, et al.. (2026). See the (In)Visible: Infrared Thermography for Characterizing Thermal Patterns Associated with Peripheral Arterial Disease and Diabetes Mellitus.. Medical sciences (Basel, Switzerland). https://doi.org/10.3390/medsci14040487