The purposes of this study were 1) to provide dermatologists as well as general practitioners with a computer aided device in order to produce an objective risk level of the melanocytic lesions helping the clinician during the diagnostic pathway 2) to compare the discrimination capacity of dermatologists and general practitioners, with the system of automated analysis ADAM (Automatic Data Analysis for Melanoma early detection).
The database contained 1212 melanocytic lesions: 54 melanomas (52 melanomas with breslow thickness <1 mm) and 1157 melanocytic nevi. The images have been captured with a videodermoscope using polarized light videocap, dotated with a videocamera scalar at a 20fold magnification (ds medica, milano).
The images have been elaborated using the processing score of adam
A training set of 27 melanomas and 279 nevi has been randomized from the dataset.
The test set has been formed by the complement (the remaining 27 melanomas and 878 nevi), resulting in this way completely independent. In this situation (test vs training set) the best results obtained were: sensitivity 74%, specificity 72%, diagnostic accuracy 72%.
Furthermore a set of images consisting in 31 melanomas (25 thin melanomas) and 103 nevi has been selected to compare the discrimination capacity of three dermatologists with the system of automated analysis ADAM.
The entire dataset used to test the implementation of the diagnostic algorithms ADAM showed a good sensitivity and specificity performance.
Compared with the physicians, the ADAM system has shown a slightly higher diagnostic performance in terms of sensitivity and a lower one in terms of specificity.
ADAM could give further support to the dermatologist to be integrated in the specific diagnostic pathway of global anamnestic and clinical-dermatoscopic evaluation by the specialist.