Analysis of Various Textural Descriptors for Ovarian Cyst Classification
Ovarian cyst is one of the main causes of infertility. Ovarian cancers are also caused by the ovarian cyst that grows in the ovary. An ovarian cyst can be benign (non-cancerous) or malignant (cancerous). If the cyst is not diagnosed and treated at the earliest, the benign cyst may turn into malignant and can be fatal. Various image processing techniques are used to assist the clinicians to characterize the ovarian cyst using the textural descriptor. This paper reviews several textural descriptors for feature extraction like Local Binary Pattern (LBP), Local Directional Pattern (LDP) and Local Optimal Oriented Pattern (LOOP). Finally, extracted features are applied to SVM, KNN and Ensemble classifiers to compare the performance of the textural descriptors.