Smartphone-based infrared camera accurately detects diabetic foot ulcers
medwireNews: A low-cost, smartphone-based infrared (IR) camera works as well as a high-end IR camera for the detection of diabetic foot ulcers (DFU), study findings indicate.
“An advanced home assessment tool to monitor the foot in people with diabetes is desirable, and for this measurement of foot skin temperature is a promising modality,” Rob van Doremalen (University of Twente, Enschede, the Netherlands) and co-authors explain.
They acquired plantar images of both feet from 32 participants (mean age 67 years, 75% men) with a current (n=28) or recently healed (n=4) DFU using the FLIR-One IR camera (FLIR Systems, Wilsonville, Oregon, USA) attached to a Motorola XT1642 Moto G4 Plus smartphone (Motorola Mobility LLC, Chicago, Illinois, USA).
The images were compared with those of the same feet taken using a high-end FLIR-SC305 thermal camera (FLIR Systems), which was considered the gold standard. All measurements were taken during a single visit to a diabetic foot outpatient clinic.
Based on previous studies, a difference in average temperature of 1.35ºC between the entire plantar side of each foot, and a difference of 2.2ºC between prespecified regions of interest, were considered “a clinical outcome.”
The researchers report in Diabetes Research and Clinical Practice that, between the two camera systems, the intra-class correlation coefficient (ICC) values were 0.987 and 0.981 for the entire plantar foot and for nine pre-specified regions of interest combined, respectively, which are both “well above the threshold (0.9) that is considered excellent agreement,” they write.
The regions of interest were the hallux, digitus 3 and 5, metatarsophalangeal joints 1, 3, and 5, midfoot, midfoot lateral, and cuboid, for which ICC values ranged from 0.929 (digit 5) to 0.993 (metatarsophalangeal joint 3).
Bland-Altman plots also showed a “negligible” clinical difference between the two cameras, with the mean difference at 0.14 for the whole plantar foot and 0.06 for the combined regions.
Compared with the high-end IR camera, the smartphone-based IR camera had sensitivities of 94% for the entire foot and 93% for the combined individual regions, and corresponding specificities of 86% and 91%.
The positive and negative likelihood ratios, which van Doremalen et al say are “the most important for clinical decision-making,” were 6.56 and 0.07, respectively, for the entire plantar foot, and 10.86 and 0.07, respectively, for the combined regions.
The positive values fell above the cutoff of 5, indicating “strong evidence” of a clinical outcome, while the negative values fell below the cutoff of 0.1, pointing to “convincing evidence.”
The investigators suggest that the smartphone-based IR camera should now be validated in patients with no developing ulcers or feet that were ulcer-free for longer periods of time.
They add that “[t]he next step is to test the predictive value of IR-cameras” in people’s homes, as well as creating “specific acquisition and automatic assessment algorithms for the smartphone application to assess the IR images.”
By Laura Cowen
medwireNews is an independent medical news service provided by Springer Healthcare. © 2019 Springer Healthcare part of the Springer Nature group