MODIFIED WHALE OPTIMIZATION ALGORITHM AND MINIMUM CROSS ENTROPY BASED SEGMENTATION OF CT LIVER IMAGE

Abstract

In the modern healthcare system, the segmentation process is really helpful to aid in the diagnosis process.  In this paper, the efficient metaheuristic approach - Modified Whale Optimization Algorithm and Minimum Cross Entropy (MWOA & MCE ) based multilevel thresholding is proposed for segmentation of the computer tomography (CT) liver image. Segmentation of liver cyst image is the main objective of this paper, which will assist doctors to diagnose liver cysts. The results of segmentation of CT liver images are compared with other algorithms, Teaching-Learning-based optimization algorithm, Jaya algorithm and Genetic algorithm, to evaluate the efficiency of the MCE & MWOA approach. Different performance calculating methods like Uniformity, Structure Similarity Index (SSIM), Root Mean Square Error (RMS Error), Rand Index (RI), Execution time, Variation of Information (VoI) and Peak Signal-to-Noise Ratio (PSNR) are assessed from an original and resultant image. In comparison to other algorithms, the proposed method's findings demonstrate that the MCE & MWOA algorithm achieve accurate and efficient segmentation results.