Spectral MD Holdings Ltd (AIM:SMD) said its AI-driven burns imaging technology was shown to be 92% accurate (up from 91%) in an expanded proof-of-concept study.
Early clinical results from the evaluation involving 124 people also revealed the DeepView wound imaging solution was 88% accurate when used in paediatric patients.
The platform, which combines state-of-the-art imaging with the latest machine learning algorithm, is effectively able to help physicians triage burn victims by predicting whether their wound will simply require dressing or surgery to heal.
Why is this important? Well, it saves time and money – and, for the patient, it can prevent a protracted and painful convalescence.
Spectral reckons the wait-and-see approach of medical staff takes up to 21 days, where DeepView’s assessment occurs in seconds.
Chief executive Wensheng Fan said: "The results not only demonstrate excellent AI performance metrics, but they also show the technology is responding well to demographic and geographic variability in the study population.
“As such, we have already bolstered our infrastructure to facilitate the expansion of the study to additional sites and have begun enrolling in a larger study to complete AI algorithm development.”
The group is developing DeepView for federal use in burns in the US and has received non-dilutive funding from the Biomedical Advanced Research and Development Authority (BARDA).
A version has also been created for diabetic foot ulcers.