EVALUATION OF DIFFERENT IMAGE ACQUISITION TECHNIQUES FOR A COMPUTER VISION SYSTEM IN THE DIAGNOSIS OF MALIGNANT-MELANOMA

Background: Digital image analysis was found to be a useful technique for improved accuracy of preoperative diagnosis of melanocytic lesions . In previous studies digitized color slides were used as input for di gital image analysis. New technologies and smaller video cameras made it possible to develop a camera system that allows the digitization of skin lesions directly from the patient.Objective We investigated whet her conventional color slides or directly digitized images should be u sed for a reliable recognition of malignant melanoma. Methods: Compute r features describing characteristics of the lesions were computed for 404 digitized color slides and for 309 directly acquired lesions. Sta tistical analysis and classifier construction was performed by the com mercial statistical classification program CART. Results: With the dat a set derived either from the color slides or from the directly digiti zed lesions a sensitivity of about 90% for the recognition of malignan t melanoma could be obtained. Conclusion: Both image acquisition techn iques allow a reliable detection of malignant melanoma and both are ap propriate as input for an image analysis system regarding its efficien cy as a diagnostic tool. However, none of the classifiers can be appli ed with reasonable significance to both techniques.