Diagnosis of human oligodendrogliomas with the help of the NeuroShell EasyClassifier (TM) neural network

OBJECTIVE: To examine whether a suitable solution can be found concerning t he ability to reproduce the histologic classification of human oligodendrog liomas with the assistance of the NeuroShell Easy Classifier(TM) neural net work. STUDY DESIGN: Histologic sections of 449 human oligodendrogliomas were sele cted. The diagnostic task was given by differentiation of three oligodendro glioma types: 121 low grade oligodendrogliomas, World Health Organization g rade 2; 180 low grade oligoastrocytomas; and 148 anaplastic oligodendroglio mas, grade 3. Age, sex and 50 histologic characteristics were examined in e ach case, describing the presence of a specific histologic feature on a sca le of four (zero, absence of the feature; three, abundant presence). From e ach group, two-thirds of randomly selected tumors were available for the tr aining set and one-third for the testing set. RESULTS: In the three-class problem, 98.88% of the tumors were correctly cl assified (testing set). Ninety-nine percent of new testing tumors were corr ectly classified with Easy Classifier(TM) as low grade and anaplastic oligo dendrogliomas. In the case of low grade oligodendrogliomas versus low grade oligoastrocytomas, 99% of new tumors were correctly classified. CONCLUSION: The main conclusion from this study is that Easy Classifier(TM) was able to differentiate, with high accuracy, sensitivity and specificity , among the three types of oligodendrogliomas.