Thermal images taken in home settings at week 0 can differentiate between healed and unhealed wounds on venous leg ulcers, according to a study published online Oct. 26 in Scientific Reports .

Quoc Cuong Ngo, Ph.D., from RMIT University in Melbourne, Australia, and colleagues used wound data for 56 older participants with venous leg ulcers over 12 weeks. Thermal images of the wounds were collected in their homes and labeled as healed or not healed. Textural information was extracted from the thermal images at week 0, with higher variation of gray tones distribution for thermal images of unhealed wounds.

The researchers noted that the first three principal components of the textural features from one timepoint can be used as an input to a Bayesian neural network, which can discriminate healed from unhealed wounds. The classification results showed 78.57 and 60.00 percent sensitivity and specificity, respectively, using the optimal Bayesian neural network.

"Our new work that identifies chronic leg wounds during the first visit is a world-first achievement," a coauthor said in a statement. "This means that specialized treatment for slow-healing leg ulcers can begin up to four weeks earlier than the current gold standard."

Abstract/Full Text