Validation of the MIA and MSKCC Nomograms for Sentinel Node Positivity Prediction in a Spanish Melanoma Cohort

Background: Skin examination through the Vectra WB360 imaging system promises to increase the diagnostic accuracy of melanoma and non-melanoma skin cancers, by early recognition of atypical lesions or detecion of changes within pre-existing ones.The purpose of our study is to assess the performance of the clinical AI algorithm integrated into the VECTRA WB360 examination system in detecting melanomas in high risk patients.Methods: A WB360 scan is routinely taken at each dermatological visit of patients at high risk for melanoma (i.e.dysplastic nevus syndrome, family or personal history of melanoma).The clinical AI algorithm is used to identify atypical melanocitic lesions by punctual macroscopic pattern analisys as well as comparison between two different acquisitions in time.Variables such as: numer of patients' nevi (range), location and clinical characteristics of atypical melanocitic lesions were recorded.A histopath correlation is then performed.Results: The clinical AI algorithm integrated in the VECTRA WB360 showed to impact on dermatologists' diagnostic accuracy, with different outcome according to the variable under assessment.Conclusions: The clinical AI algorithm integrated into the vectra WB360 imaging system is helpful in monitoring patients at high-risk of melanoma.