Artificial neural network for cervical abnormalities detection on computed tomography images

<span lang="EN-US">Cervical cancer is the second deadliest after breast in Indonesia. Sundry diagnostic imaging modalities had been used to decide location and severity of cervical cancer, one among those computed tomography (CT) Scan. This study handles a CT image dataset consisting two categories, abnormal images patients normal cervix with other diseases. It focuses on ability segmentation classification programs localize areas classify into categories based features contained them. We conferred novel methodology for contour detection round organ classified artificial neural network (ANN) which was employed categorize data. The algorithm region-based snake model. texture area were arranged form gray level co-occurrence matrix (GLCM). Support vector machine (SVM) added determine better comparison. Experimental results show that ANN model has receiver operating characteristic (ROC) parameter values than SVM model’s existing approach’s regarding 96.2% sensitivity, 95.32% specificity, 95.75% accuracy. </span>