456 Multispectral imaging algorithm predicts Breslow tumor depth of malignant melanoma

Breslow tumor thickness is a major prognostic factor for melanoma. As it based on histopathological evaluation, not available to aid clinical decision making at the time of melanoma diagnosis. In this work, we assessed efficacy multispectral imaging (MSI) predict and developed classification algorithm determine optimal surgical safety margins. First, excluded nevi from analysis with use novel quantitative parameter. Based parameter, could differentiate melanomas sensitivity 89.60% specificity 88.11%. This initial step was followed by categorization into three different subgroups (≤1 mm, 1-2 mm >2 mm) 78.00% 89.00% substantial agreement (κ = 0.67; 95% CI, 0.58-0.76). We compared our results performance dermatologists dermatology residents who dermoscopic images these reached 60.38% 80.86% moderate 0.41; 0.39-0.43). results, method may prediction appropriate margins curative excision.