A Computer Aided Diagnostic Tool for the Detection of Uterine Fibroids

Abstract

The integration of information technology with biomedicine has provided viable diagnostic tools to the medical community. Such computer aided procedures fastens the clinical decision process without any hurdle. Among different medical imaging modalities, Ultrasonic Imaging plays a vital role in detecting gynecological pathologies. Of importance, Uterine fibroid detection requires significant attention where symptoms such as, infertility and miscarriage can be predicted. This paper suggests an automated computer aided diagnostic tool for the detection of uterine fibroid. Gabor wavelets are applied for texture segmentation and statistical features such as mean, variance, standard deviation, skewness, kurtosis, Eigen values, GLCM contrast and energy are extracted from the user defined region of interest (ROI). The qualitative procedure is examined using the morphological operations and gray level intensity variations. Two neural network models, multilayer perceptron neural network (MLP) and probabilistic neural network (PNN) are applied to classify the normal and fibroid uterus image. It is found from the experimental computer simulation, a classification accuracy of 97.25% is obtained using combinational statistical features, mean and standard deviation with PNN classifier. It can be concluded that the proposed tool can applied as an efficient Medical Expert System for diagnosing the Ultrasonic Uterus images.