FRACTAL AND INTEGER-DIMENSIONAL GEOMETRIC ANALYSIS OF PIGMENTED SKIN-LESIONS

Accurate in vivo diagnosis of pigmented skin lesions is required to id entify and excise malignant melanomas but to avoid unnecessary excisio n of benign lesions; the published rates of clinical diagnostic accura cy are about 65%. This study investigates whether fractal geometric an alysis of pigmented skin lesions can improve the rate of diagnostic ac curacy. Forty-two pigmented skin lesions (15 malignant melanomas, 21 m elanocytic naevi, and 6 basal cell papillomas) on patients attending a dermatology clinic were photographed, excised, and sent for histopath ological examination. The fractal dimension of the boundary of the les ions was measured using a box-counting method implemented on a microco mputer-based image analysis system. Euclidean geometric parameters wer e also measured. The fractal dimension of all the lesions was greater than the topological dimension (one), indicating that there is a fract al element to their structure. Using all measured parameters together, multivariate linear discriminant analysis produced a confusion matrix in which 45% of the lesions were assigned to the correct diagnostic g roup with a kappa statistic of 0.33. There was no significant differen ce between the fractal dimension of melanocytic naevi and that of mali gnant melanomas (p = 0.18), Although pigmented skin lesions have a fra ctal element to their structure, the fractal dimension of their bounda ries is not a useful morphometric discriminant between the diagnostic groups of malignant melanomas and benign melanocytic naevi.