Efficient Tumor Segmentation in Liver using Spatial Fuzzy C Means Clustering with Enhanced MUMFORDSHAH Approach

Abstract

Liver tumor is an abnormal growth of tissues which develops in the liver with a growing morbidity rate. To maximize the rate of survival, early detection and effective treatment are vital in which medical imaging tools can assist tremendously. Computed Tomography (CT) is one of several imaging techniques used to detect hepatic illness since it may offer relatively high-resolution pictures with exact anatomical details. Analyzing huge number of CT scans, on the other hand, is a massive undertaking for radiologists. Due to the necessity for precise and efficient tumor identification, semiautomatic and automatic tumor segmentation methodologies have been introduced. In this article based on the Spatial FCM (SFCM) with Enhanced Mumford Shah methodology, an efficient automatic segmentation approach for segmenting liver tumor in CT volumes was developed. Here, Mumford shah approach is improvised to enhance the robustness of the process while consuming less time. When compared to existing algorithms such as K-means and Fuzzy C- Mean (FCM), the suggested approach outperformed in terms of accuracy, specificity, and sensitivity.