Study: Skin cancer detection apps ‘cannot be relied on’ for an accurate diagnosis

While recognising their potential, the authors have warned against their use in the current state.
By Sophie Porter
Share

A study published in the BMJ has advised against the use of apps to assess the risk of skin cancer.

Primarily targeted at catching early-stage melanoma by photographing suspicious skin lesions on a smartphone, the apps were found to be subject to “poor and variable performance”, in spite of one having gained CE marking.

The discrepancy has led the authors to condemn the European regulatory process, stating the current system “does not provide adequate protection to the public”.

WHAT HAPPENED

The study focused on nine studies of different downloadable algorithm-based smartphone apps, including the widely available SkinVision. It found a scarcity of reliable data supporting the efficacy of the apps, with the studies exhibiting selective participant recruitment, a lack of consistent classifications and inadequate reference standards.

For instance, in five of the studies analysed, clinicians chose which skin lesion to test, and in seven studies, researchers rather than participants photographed them. The authors said the control of layman participation likely skewed the results away from real world usage.

WHY IT MATTERS

The availability and convenience of the apps may cause people to use them in lieu of seeking professional medical advice, which could result in undiagnosed melanoma or unnecessary anxiety in patients who have received a false positive.

The authors highlight the “concern (...) about the impact of false reassurances” as the apps themselves market this trust, with SkinVision claiming to offer “a complete skin cancer detection service (...) to make sure you stay safe”, whilst lacking the adequate performance to back it up.

The authors also say that the efficacy of the app is contingent on their limited resource of identifiable skin conditions, the quality of the smartphone camera and the initiative of patients in contacting medical professionals upon the receipt of a ‘moderate risk’ result.

However, in the study, they mistakenly refer to an app called skinScan from TeleSkin, based on a study from 2014, which they now understand was in fact looking at an earlier version of SkinVision. 

“We apologise to TeleSkin for this error,” the authors said yesterday. “The review currently contains no data on the accuracy of the any version of the skinScan app from TeleSkin. We have been made aware of an unpublished study of the TeleSkin skinScan app which we will endeavour to add to the review.”

THE LARGER PICTURE

On average, one in five people will develop a form of skin cancer during their lifetime, with incidence increasing every year. Doctors rely on early detection as the most efficient treatment, particularly with melanoma, which has up to a 95% five-year survival rate if identified early.

Health tech is being used increasingly to aid in early detection. For instance, in 2019, SkinVision was integrated into the NHS.

The authors of the study acknowledge the potential of this kind of apps, but warn against their use in the current state, encouraging more rigorous testing and development before their approval and adoption.

ON THE RECORD

“In a rapidly advancing field, quality of evidence is poor to support the use of these apps to assess skin cancer risk in adults with concerns about new or changing skin lesions,” they write. “The current CE marking assessment processes are inadequate for protecting the public against the risks created by using smartphone diagnostic or risk stratification.”

Zeljko Ratkaj, chief executive of TeleSkin, told MobiHealthNews: “I totally agree that this area, mobile application usage in healthcare, needs to be regulated more strictly. It also needs to be very clear what is the extent of the app usage and what the applications are and what are not.

“However, the process of certification is really hard and, for young startups, can be an impossible obstacle to cross, due to the funding limits (process costs), time and effort that needs [sic] to be put in.”

Ratkaj said that while he supported efforts to strengthen regulation, he would also like to see more collaboration between medical institutions and health tech companies.

In a statement, SkinVision said: “We believe that for the benefit of public health, all algorithm-based smartphone apps should be researched and would like to continue leading by example, as we continue supporting clinical research of our algorithm. 

“While there are risks associated with all medical devices, research also shows that the use of algorithm-based apps alongside healthcare professionals can provide a greater benefit for patients and the healthcare system. We have assisted in finding over 40,000 cases of skin cancer already. These objective facts prove that the clinical benefits of the service outweigh the risks.”