Detection of Ovarian tumour by soft computing techniques

Abstract

Ovarian tumour originates in ovaries posing a serious threat to newlinewomen. As a result, it leads to abnormal cells which have the ability to newlinespread to the other regions of the body. Ovarian tumour is a sort of risky newlineimprovement that impacts ovaries in females, and is hard to perceive at newlineearly the phase because of which it stays as one of the guideline newlinewellsprings of illness end. Unquestionable confirmation of intrinsic and newlinetypical parts is immense in making novel frameworks to perceive and newlineruin danger. The ultrasound imaging is the low cost and best to detect newlinetumour in ovaries. To improve tumour detection accuracy in ultrasound newlineovarian images, enhanced preprocessing, image segmentation, feature newlineextraction and feature selection, and improved classification method are newlineproposed in this research work. newlineAn Improved Anisotropic Diffusion Filter (IADF) is proposed newlineinitially to remove unnecessary parts and noises from the ultrasound newlineovarian tumour image for the image quality improvement. Afterward, newlineImproved Whale Search Optimization (IWSO) is proposed for the image newlinesegmentation process to segment tumour in ultrasound images newlineeffectively. An Improved Ant Swarm Optimization (IASO) algorithm is newlineproposed to process the feature extraction and feature selection. As a newlinevi newlineresult, optimal features are extracted and chosen by IASO from the newlineultrasound ovarian tumour image for the further classification process. newlineFinally, Hybrid Swallow Swarm Intelligence-Deep Neural Network newline(HSSI-DNN) classification method is applied to classify the tumour in newlinethe ultrasound image as benign or malignant based on the selected newlinefeatures. newline