Image processing in clinical decision support system

Cervical cancer is the second most common cancer in women world-wide. The accuracy of colposcopy is highly dependent on the physicians individual skills. In expert hands, colposcopy has been reported to have a high sensitivity (96%) and a low specificity (48%) when differentiating abnormal tissues. This leads to a significant interest to activities aimed at the new diagnostic systems and new automatic methods of coloposcopic images analysis development. The presented paper is devoted to developing method based on analyses fluorescents images obtained with different excitation wavelength. Our approach involves images acquisition, image processing, features extraction, selection of the most informative features and the most informative image types, classification and pathology map creation. The special preprocessing procedures: automatic regions of interest (ROI) segmentation and multispectral fluorescent images matching were realized for each image set. The classification strategy is RDF Random Decision Forest. The result of proposed method is pathology map creation - the image of cervix shattered on the areas with the definite diagnosis such as norm, CNI (chronic nonspecific inflammation), CIN(cervical intraepithelial neoplasia). The obtained result of experimental investigation on the border CNI/CIN: sensitivity - 0.85, the specificity -0.78. A proposed algorithm gives possibility to obtain correct differential pathology map with probability 0.8. The figures of sensitivity and specificity correspond to estimates of sensitivity and specificity for colposcopic examination conducted by an experienced physician and exceeds characteristics of inexperienced physician.