Computerized system to enhance the clinical diagnosis of pigmented cutaneous malignancies.

BACKGROUND: An increase in the incidence of cutaneous malignant melanoma in recent years has not been accompanied by satisfactory progress in diagnostic methods. This study was carried out to evaluate a specially designed computerized image analysis system, called Derma Vision, to aid in the differentiation between malignant and benign cutaneous pigmented lesions.
METHODS: Seventy-one lesions were photographed with a digital camera and the data were analyzed by the Derma Vision system. The system assessed the variation of hues in each image, calculated the mean standard deviation of the hues, and produced a value that expressed the range of hues in the lesion. The lesions were then excised and examined histologically. The computer-assisted clinical diagnosis was correlated with the histologic diagnosis to determine the accuracy of the former.
RESULTS: Derma Vision predicted the malignant character of a lesion with 92% precision, compared with 87% accuracy based only on the clinical features.
CONCLUSIONS: This simple, inexpensive device can boost the accuracy of clinical diagnosis and provide a useful tool to the physician faced increasingly with having to determine whether pigmented lesions are malignant or benign.