HealthDay News — Current smartphone-based applications (apps) cannot detect all cases of melanoma or other skin cancers, according to a review published online Feb. 10 in The BMJ.
Karoline Freeman, from the University of Birmingham in the United Kingdom, and colleagues conducted a systematic review to examine the validity and findings of studies that examined the accuracy of algorithm-based smartphone apps to assess the risk for skin cancer. Nine studies that assessed six different apps were included.
The researchers note that the studies were small and of poor methodological quality; recruitment was selective, and there were high rates of unevaluable images and differential verification. Clinicians rather than smartphone users performed lesion selection and image acquisition. Two of the apps were available for download. SkinScan was assessed in a single study with 15 lesions and five melanomas and was found to have 0 and 100 percent sensitivity and specificity, respectively, for melanoma detection. SkinVision was assessed in two studies with 252 lesions and 61 malignant or premalignant lesions and achieved sensitivity and specificity of 80 and 78 percent, respectively, for detection of malignant or premalignant lesions. In three studies, the accuracy of SkinVision was poor when verified against expert recommendations.
“As technologies continue to develop, these types of apps are likely to attract increasing attention for skin cancer diagnosis, so it’s really important that they are properly evaluated and regulated,” one coauthor said in a statement.
One author disclosed financial ties to a technology company and the publishing industry.