QUANTITATIVE, MICROSCOPICAL, COMPUTER-AIDED DIAGNOSIS OF ENDOMETRIAL HYPERPLASIA OR CARCINOMA IN INDIVIDUAL PATIENTS

There are many significant differences but also considerable overlap among the quantitative histopathological features of mild and marked atypical endometrial hyperplasias and well and moderately differentiated carcinomas, thus preventing its application to individual patient care. To overcome this problem, a classification rule for the diagnosis in individual patients using discriminant analysis was developed. Utilizing 9 quantitative features, the above 4 groups can be adequately separated. None of the carcinomas was misclassified as hyperplasia and only 1 case of marked atypical hyperplasia was erroneously classified as well-differentiated carcinoma, with a probability for carcinoma of 0.75, hyperplasia 0.25. The classification probabilities of all the confirmed carcinomas exceeded 0.90. Using 0.90 as a classification level (threshold) a reliable rule is obtained. A slightly more simple classification rule distinguished among all the hyperplasias and all the carcinomas and gave correct classification in all cases of the test set. The classification rules can be used to select patients with benign disease for hormone therapy as an alternative to hysterectomy and can be programmed into an inexpensive microcomputer.