Texture-Based Classification of Hysteroscopy Images of the Endometrium

Abstract

The objective of this study was to classify hysteroscopy images of the endometrium based on texture analysis for the early detection of gynaecological cancer. A total of 418 regions of interest (ROIs) were extracted (209 normal and 209 abnormal) from 40 subjects. Images were gamma corrected and were converted to gray scale. The following texture features were extracted: (i) statistical features, (ii) spatial gray level dependence matrices (SGLDM), and (iii) gray level difference statistics (GLDS). The PNN and SVM neural network classifiers were also investigated for classifying normal and abnormal ROIs. Results show that there is significant difference (using Wilcoxon rank sum test at a=0.05) between the texture features of normal and abnormal ROIs for both the gamma corrected and uncorrected images. Abnormal ROIs had lower gray scale median and homogeneity values, and higher entropy and contrast values when compared to the normal ROIs. The highest percentage of correct classifications score was 77% and was achieved for the SVM models trained with the SF and GLDS features. Concluding, texture features provide useful information differentiating between normal and abnormal ROIs of the endometrium.