Evidence assessing the diagnostic performance of medical smartphone apps: a systematic review and exploratory meta-analysis

Abstract

ObjectiveThe number of mobile applications addressing health topics is increasing. Whether these apps underwent scientific evaluation is unclear. We comprehensively assessed papers investigating the diagnostic value of available diagnostic health applications using inbuilt smartphone sensors.MethodsSystematic Review—MEDLINE, Scopus, Web of Science inclusive Medical Informatics and Business Source Premier (by citation of reference) were searched from inception until 15 December 2016. Checking of reference lists of review articles and of included articles complemented electronic searches. We included all studies investigating a health application that used inbuilt sensors of a smartphone for diagnosis of disease. The methodological quality of 11 studies used in an exploratory meta-analysis was assessed with the Quality Assessment of Diagnostic Accuracy Studies 2 tool and the reporting quality with the ’STAndards for the Reporting of Diagnostic accuracy studies' (STARD) statement. Sensitivity and specificity of studies reporting two-by-two tables were calculated and summarised.ResultsWe screened 3296 references for eligibility. Eleven studies, most of them assessing melanoma screening apps, reported 17 two-by-two tables. Quality assessment revealed high risk of bias in all studies. Included papers studied 1048 subjects (758 with the target conditions and 290 healthy volunteers). Overall, the summary estimate for sensitivity was 0.82 (95 % CI 0.56 to 0.94) and 0.89 (95 %CI 0.70 to 0.97) for specificity.ConclusionsThe diagnostic evidence of available health apps on Apple’s and Google’s app stores is scarce. Consumers and healthcare professionals should be aware of this when using or recommending them.PROSPERO registration number42016033049.

Deep‐learning‐based, computer‐aided classifier developed with a small dataset of clinical images surpasses board‐certified dermatologists in skin tumour diagnosis

“…There are over 165 000 healthcare mobile applications available on the market, some of which provide the probability or risk of photographed melanocytic skin conditions being benign or malignant . Buechi et al .…”

“…There are over 165 000 healthcare mobile applications available on the market, some of which provide the probability or risk of photographed melanocytic skin conditions being benign or malignant . Buechi et al . conducted a meta‐analysis of smartphone applications and reported that the pooled sensitivity and specificity of studies assessing melanoma were 73% and 84%, respectively .…”

Improving the early diagnosis of early nodular melanoma: can we do better?

“…[50] Some authors demonstrate that these applications have a sensitivity for melanoma of 73% and specificity of 83%, whereas dermatologists have sensitivity of 88% and specificity of 97%. [51] Other authors demonstrate that smartphone apps have scarce scientific evidence, studies are small and have low methodological quality and only one-third of the app has assessed diagnostic accuracy [52]. Thus, new apps may come to the market, which may significantly aid the diagnosis of early NM.…”

The effect of active occupational stress management on psychosocial and physiological wellbeing: a pilot study

“…Research with diagnostic studies are missing since most apps target a diagnostic problem without the scienti c basis. 22 Our study is a good example of how a mHealth solution might be integrated into the occupational health programs (OHP) and prevention programs. The results showed signi cant changes in levels of stress hormones which could be due to increased level of physical activity and frequency of deep relaxation during a regular week for employees.…”