Performance of smartphone applications in assessing melanoma risk is highly variable, with three of four applications incorrectly classifying 30% or more of melanomas as unconcerning, according to a study.
To measure the performance of smartphone applications that evaluate photographs of skin lesions and provide the user with feedback about the likelihood of malignancy, Joel A. Wolf and colleagues at the University of Pittsburgh Medical Center tested the sensitivity, specificity and positive and negative predictive values of four smartphone applications.
The authors included 188 images of lesions in the analysis, evaluating each using the four smartphone applications and recording the test result as positive, negative or not able to be evaluated. Of these lesions, 60 were melanoma and the remaining 128 were benign.
Sensitivity (the proportion of actual positives correctly identified) of the four applications tested ranged from 6.8% to 98.1%; specificity (the proportion of actual negatives correctly identified) ranged from 30.4% to 93.7%; positive predictive value (the ratio of true positives to combined true and false positives) ranged from 33.3% to 42.1%; and negative predictive value ranged from 65.4% to 97%.
The highest sensitivity for melanoma diagnosis was observed for an application that sends the image directly to a board-certified dermatologist for analysis, while applications with the lowest sensitivity for melanoma diagnosis used automated algorithms to analyze images.
The authors suggested that reliance on these applications, which are not subject to regulatory oversight, and not seeking medical consultation can delay the diagnosis of melanoma and potentially harm users.
“Physicians must be aware of these applications because the use of medical applications seems to be increasing over time,” the authors concluded. “The dermatologist should be aware of those [applications] relevant to our field to aid us in protecting and educating our patients.”
The study is scheduled for publication in JAMA Dermatology (formerly the Archives of Dermatology), and is available at http://archderm.jamanetwork.com/article.aspx?articleid=1557488#qundefined.