Apps to diagnose skin cancer are unreliable, warn researchers
DOCTORS have issued a warning against using smartphone apps to diagnose moles as skin cancer after three out of four tested wrongly reassured patients.
Smartphone apps are increasingly being used in healthcare and some have been validated by regulators or medical authorities.
However a selection of unvalidated apps designed to check if a skin lesion is concerning or not were unreliable, researchers said.
They choose four unidentified apps which analysed photographs of skin lesions and gave responses such as 'melanoma', or 'problematic' or 'low risk'.
Three out of the four did not directly involve doctors and used automated programmes to assess the picture.
The researchers from the University of Pittsburgh found the three automated apps incorrectly classified 30 per cent or more melanomas as unconcerning.
This meant patients would have been at risk of developing the most serious form of skin cancer yet had been wrongly reassured the lesion was benign.
The findings were published in the Journal of the American Medical Association Dermatology.
Joel Wolf, and colleagues at the University of Pittsburgh Medical Center, uploaded 188 images of lesions in the analysis, each of which was evaluated by the four smartphone applications, and the test result was recorded as positive, negative or unevaluable. Of these lesions, 60 were melanoma and the remaining 128 were benign.
The apps correctly identified the skin cancers in between seven per cent of cases and 98 per cent of cases.
The app that sent images to a certified dermatologist for analysis performed the best with melanomas correctly identified 98 per cent of the time.
Even in the best performing app of the three that did not involve a doctor 18 of the 60 melanoma skin cancers were said to be benign.
The app using a dermatologist incorrectly identified one melanoma from 53 as benign. This app was also the most costly, at $5 per image where as the automated apps varied in cost from free to $4.99 for an unlimited number of images.
Mr Wolfe wrote: "Technologies that improve the rate of melanoma self-detection have potential to improve mortality due to melanoma and would be welcome additions to our efforts to decrease mortality.
"However, extreme care must be taken to avoid harming patients in the process.
"Physicians must be aware of these applications because the use of medical applications seems to be increasing over time.
"Despite disclaimers that these applications are intended for educational purposes, they have the potential to harm users who may believe mistakenly that the evaluation given by such an application is a substitute for medical advice."