Panos Papachristou

Background: Artificial intelligence has in numerous studies shown high accuracy in detecting skin cancer when trained on various databases of dermatoscopic images. However, there are very few prospective studies conducted in real clinical settings directed at patients seeking healthcare for assessment of skin lesions, and no such studies at all in primary care, where the majority of patients are managed.Project aim: To study the accuracy, reliability, and clinical utility of an AI-based decision support system (Dermalyser), developed for primary care, in distinguishing skin cancer from benign lesions.Method: Randomized controlled trial at 30 primary care centers in Sweden, Germany, and the Netherlands, with consecutive inclusion of patients (n=2000) seeking medical evaluation of skin lesions of concern. Participants will be equally randomized to either an intervention group, where the examining physician (if applicable) uses the Dermalyser in the clinical assessment, or to a control group where the assessment follows standard clinical routine, without involvement of Dermalyser. Both groups will be followed for up to 5 years, with regard to the tumour diagnoses, proportions of skin cancer/benign lesions, and morbidity and mortality in skin cancer. Possible between-group differences will be investigated statistically. Befits: If the Dermalyser prooves to be safe and diagnotically reliability, it could enhance the chance of detecting skin cancer in early stage in primary care.