Thursday, April 18, 2013 (Last Updated: 04/19/2013)
Joel A. Wolf, from the University of Pittsburgh, and colleagues compared the diagnostic accuracy of four smartphone applications that evaluate photographs of skin lesions and provide feedback about the likelihood of malignancy with a histologic diagnosis from a dermatopathologist. Sixty melanomas and 128 benign control lesions were assessed.
The researchers found that sensitivity ranged from 6.8 to 98.1 percent; specificity ranged from 30.4 to 93.7 percent; positive predictive value ranged from 33.3 to 42.1 percent; and negative predictive value ranged from 65.4 to 97.0 percent. An application that sent the image to a dermatologist for analysis was most sensitive, while applications that analyzed images with automated algorithms were least sensitive.
"The performance of smartphone applications in assessing melanoma risk is highly variable, and three of four smartphone applications incorrectly classified 30 percent or more of melanomas as unconcerning," Wolf and colleagues write.
One author disclosed being an investigator and consultant for MELA Sciences.
Hematology & Oncology