The segmentation of liver images from computed tomography (CT) scans is a pivotal technique that supports various medical applications, including computer -aided diagnostics, disease identification, and the evaluation of hepatic function. In this study, an advanced segmentation method for CT liver images is introduced, leveraging the synergy between Renyi entropy and fuzzy c -partition methodologies. The proposed approach commences with the enhancement of input CT images employing an adaptive histogram equalization technique, thereby improving the contrast of hepatic tissues. Subsequently, these images are transformed into the fuzzy domain, wherein the entropies of the hepatic object and the surrounding tissue are meticulously defined. The optimization of the Renyi entropy measure is adeptly carried out using the Differential Evolution (DE) algorithm, which establishes precise CT image thresholds for segmentation. The efficacy of the proposed framework is substantiated through extensive experiments, which reveal its superior performance in segmenting liver CT images against complex backgrounds. The results affirm the framework's proficiency, particularly in medical imaging contexts with intricate backdrops, thereby underscoring its potential for enhanced diagnosis and therapeutic planning.
原文链接
分享
引用
