Skin cancer is the most common cancer in the US, accounting for almost half of all cancers. But if it is detected early – before it has spread to other parts of the body – it is almost always curable. There are now an array of smartphone applications available claiming to aid early detection of skin cancer. But how effective are they?
Each year, more than 3.5 million cases of basal and squamous cell skin cancers are diagnosed in the US, and last year saw the diagnosis of over 76,000 cases of melanoma – the most deadly form of the disease.
Basal and squamous cell skin cancers, also referred to as non-melanoma skin cancers, are those that develop on the outer layers of the skin – in the basal cells or squamous cells. These are the most common form of skin cancers.
Melanoma is a skin cancer that forms in the cells responsible for skin pigmentation, known as melanocytes. Although it is less common than basal or squamous cell skin cancers, it is much more aggressive. Of the 13,000 skin cancer deaths annually, melanoma accounts for around 9,700.
Symptoms of both non-melanoma and melanoma skin cancers include a change in the size or color of a mole, growth or spot on the skin, a new growth, a change in skin sensation – such as itchiness, pain or tenderness – and a spread of pigmentation outside the border of a mole or mark on the skin.
A number of smartphone apps claim to be able to assess certain skin changes and inform individuals whether such changes warrant a visit to a dermatologist for further analysis.
But how accurate are these apps? Do they aid early detection of skin cancer? Or are they putting patients at risk by delaying diagnosis? We investigate.
The rise of skin cancer apps
In 2012, there was report on the creation of a skin cancer app called UMSkinCheck.
The app – available to download for free from the iPhone App Store – provides guidance on how to check for skin lesions and moles, and users can take photos of any suspicious lesions or moles and have the app assess them. The app also reminds users to monitor any skin changes over time, and it includes information on skin cancer prevention.
Last year, there was another report on skin cancer app called Mole Detective – available for purchase through Google Play. The app analyzes pictures of skin moles and analyzes them using the ABCDE (asymmetry, border, color, diameter and evolution) method undertaken by dermatologists. The app then calculates a person’s risk of skin cancer dependent on the characteristics of their mole.
Most recently came the launch of an app called SkinVision, which claims to assist in the early detection of melanoma.
SkinVision uses a mathematical theory called “fractal geometry” to analyze photos of skin lesions and moles taken by the user. According to the manufacturers, SkinVision is the first app to use this theory for early melanoma detection.
The video below provides more information about how SkinVision works:
<iframe width=”734″ height=”444″ src=”//www.youtube.com/embed/uBYOwH8N3Lw” frameborder=”0″ allowfullscreen></iframe>
In a study published in the Journal of the European Academy of Dermatology and Venereology last year, researchers tested the accuracy of the app through the analysis of 195 pictures of skin lesions. The app was 81% accurate in detecting melanoma, with 73% sensitivity and 83% specificity. However, clinical diagnoses through dermatologists presented 88% sensitivity and 97% specificity.
“With an accuracy of 81% in detecting melanoma, the algorithm might have some potential in the future for the evaluation of melanocytic nevi, but is to date insufﬁcient to detect melanoma accurately,” noted the researchers.
And this is not the only study to question the accuracy of skin cancer apps.
Reliance on skin cancer apps for detection ‘not yet warranted’
Last year – in a study published in JAMA Dermatology – Joel A. Wolf and colleagues from the University of Pittsburgh, PA, analyzed four smartphone apps that claim to detect skin cancer.
Three of the apps used algorithms to assess the risk of cancer among 188 photos of skin lesions – 60 pre-diagnosed as melanoma and 128 pre-diagnosed as benign – while the fourth app sent the photos to a dermatologist for assessment.
The fourth app was found to be the most accurate. The other three apps that used algorithms, however, were found to incorrectly categorize a large number of skin lesions. Even the most accurate app missed almost 30% of melanomas, diagnosing them as low-risk lesions.
Talking to MNT, Dr. Frank Perna, program director for the Division of Cancer Control and Population Sciences at the National Cancer Institute, says such results indicate that skin cancer apps should not be relied on for cancer detection:
“The published studies suggest that, at this point, there is large variability in the sensitivity and specificity of these applications for melanoma detection. As such, reliance on these applications is not warranted.”
It should be noted that the majority of skin cancer app developers point out that users should not solely rely on the apps for skin cancer detection. SkinVision, for example, state on their website:
“The content is not intended to be a substitute for professional medical advice, diagnosis or treatment. Never disregard professional medical advice or delay seeking it because of something you have read on SkinVision. Do not rely on information from SkinVision instead of seeking professional medical advice.”
Could skin cancer apps delay diagnosis?
But in these modern times, we tend to rely on technology for many aspects of life, and a 2012 survey from the Pew Research Internet Project revealed that around 35% of US adults say they have used online resources – including health apps – to diagnose a medical condition.
Of course, health apps have their uses. It is great to be able to constantly monitor blood pressure and keep track of our weight or what we are eating. But when it comes to a diagnosis of skin cancer, some professionals believe that relying on a smartphone app for detection is worrying.
“If patients are using one of these [apps] to say ‘melanoma’ or ‘not melanoma,’ they’re going to get in trouble. This can’t be a person’s only dermatologist.” Ken Beer, a dermatologist in West Palm Beach, FL, told The Wall Street Journal last year.
Dr. Darrell Rigel, of the NYU Langone Medical Center in New York, NY, echoed this comment when talking to ABC News:
“It is very concerning that these apps are used for diagnosis by patients, as it could lead to delay in diagnosis of melanoma, the cancer which is perhaps the most critical in early diagnosis being important for survival.”
The potential benefits of skin cancer apps
Although many health professionals have raised concerns regarding the reliance on skin cancer apps, others have praised such apps in terms of potentially raising awareness among the general public and encouraging patients to visit a dermatologist for assessment.
Researchers of the study that assessed the SkinVision app, for example, say:
“We see the main potential for the smartphone applications in the improvement of the patient-doctor communication by making aware of the need of skin cancer screening and by giving a basis of interaction.”
Talking to MNT, Dr. Perna notes that some skin cancer apps may also aid skin cancer prevention. SkinVision – and other apps – provide information of the ultraviolet (UV) index in a user’s area, allowing them to take appropriate action to reduce excess UV exposure – a primary cause of skin cancer.
Dr. Perna added that UV index information alone is unlikely to be enough to change a person’s UV-protective behavior, but that it could help.
“Considering that many people receive excess UV exposure through incidental exposure – such as by engaging in physical activity or occupational tasks – there may be opportunities to leverage physical activity tracking platforms/devices with platforms that provide UV index information,” he said.
“This approach remains to be tested. However, given technological developments and opportunities for leveraging mobile health, we’d like to encourage further research in this domain.”
More work is needed to improve skin cancer apps
At this point in time, health professionals are in agreement that skin cancer apps are not accurate enough to be used in replacement of a dermatologist for assessing an individual’s risk of skin cancer. But will they ever reach a point when they are?
Dr. Perna told us:
“It appears more work needs to be done to improve these applications, but it is ongoing. Ultimately, the degree to which an app-intervention approach yields poor results and possibly iatrogenic effects is an empirical research question.
When enough technical improvements and linkages with behavior-change strategies have been incorporated, I believe the data would be informative.”
But for now, health professionals and organizations worldwide recommend that if an individual notices any suspicious marks or lesions on the skin, or changes in moles or skin sensation, the first port of call should be a doctor or dermatologist.