Apps for skin cancer delay diagnosis

Consumers who use smartphone applications to analyze suspicious lesions or moles are at risk of having inaccurate diagnoses and delayed treatments, results of a new study indicate.

Researchers with University of Pittsburgh Medical Center conducted a case-control diagnostic accuracy study of four smartphone apps purportedly designed to detect melanoma. Of these, three incorrectly classified 30 percent or more melanomas, according to a news release.

The app with the highest sensitivity for melanoma diagnosis was one that sends the user-uploaded image directly to a board-certified dermatologist for analysis. Sensitivity of the apps ranged from 6.8 percent to 98.1 percent. Those with the lowest sensitivity used automated algorithms to assess the downloaded images, according to the study abstract.

Study authors noted that the popularity of healthcare-related smartphone apps is likely to rise, but that any apps offering medical advice could be dangerous to the user’s health if the advice is misleading or inaccurate.

The skin cancer detection apps reviewed in the study aren’t subject to regulatory oversight, and although they are meant only for educational purposes, study authors noted, some consumers may mistake them as a substitute for advice from a physician.

“Technologies that improve the rate of melanoma self-detection have potential to improve mortality … and would be welcome additions to our efforts to decrease mortality through early detection,” study authors concluded. “However, extreme care must be taken to avoid harming patients in the process.”

The findings were published online Jan. 16 in theJournal of the American Medical Association – Dermatology.