Researchers have concluded that the visual inspection of a suspicious skin lesion using the naked eye alone is not enough to ensure the accurate diagnosis of skin cancer.
A recent study brings together a large body of research regarding the accuracy of tests used to diagnose skin cancer, and essentially summarises research evidence assessing the accuracy of different diagnostic tests to support clinical and policy related decision making in the diagnosis of skin cancer.
The diagnosis of skin cancer
Dr Jac Dinnes, of the University of Birmingham‘s Institute of Applied Health Research, England, explains: “Early and accurate detection of all skin cancer types is essential to manage the disease and to improve survival rates in melanoma, especially given the rate of skin cancer world-wide is rising.
“The visual nature of skin cancer means that it can be detected and treated in many different ways and by a number of different types of specialists, therefore the aim of these reviews is to provide the world’s best evidence for how this endemic type of cancer should be identified and treated.
“We have found that careful consideration should be given of the technologies that could be used to make sure that skin cancers are not missed, at the same time ensuring that inappropriate referrals for specialist assessment and inappropriate excision of benign skin lesions are kept to a minimum.”
Key findings of the research
Visual inspection using the naked eye alone is not good enough and melanomas may be missed. Moreover, smartphone applications used by people with concerns about new or changing moles or other skin lesions have a high chance of missing melanomas.
When used by specialists, dermoscopy – a technique using a handheld device to zoom in on a mole and the underlying skin – is better at diagnosing melanoma than visual inspection alone and may also help in the diagnosis of basal cell carcinoma.
Dermoscopy is already widely used by dermatologists to diagnose melanoma but its use in primary care has not been widely evaluated therefore more specific research is needed. But this might also help GPs to correctly identify people with suspicious lesions who need to be seen by a specialist.
Artificial intelligence techniques, such as computer-assisted diagnosis (CAD), can identify more melanomas than doctors using dermoscopy images. However, unfortunately CAD systems also produce far more false positive diagnoses than dermoscopy and could lead to considerable increases in unnecessary surgery.
The alternative ways in which skin cancer can be diagnosed is meticulous, however further research is required to comprehensively ensure the accurate diagnosis of skin cancer.
