Polyp image segmentation based on improved planet optimization algorithm using reptile search algorithm

• Published in: Neural computing & applications Vol. 37; no. 8; pp. 6327 - 6349
• Main Authors: Abd Elaziz, Mohamed, Al-qaness, Mohammed A. A., Al-Betar, Mohammed Azmi, Ewees, Ahmed A.
• Format: Journal Article
• Language: English
• Published: London Springer London 01.03.2025; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Cancer; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Data Mining and Knowledge Discovery; Diagnosis; Health services; Image analysis; Image Processing and Computer Vision; Image segmentation; Medical imaging; Optimization; Optimization algorithms; Original Article; Performance enhancement; Polyps; Probability and Statistics in Computer Science; Reptiles; Search algorithms; Signal to noise ratio; Planet optimization algorithm; Polyp images; Reptile search algorithm; Multilevel thresholding image segmentation
• ISSN: 0941-0643; 1433-3058; 1433-3058
• DOI: 10.1007/s00521-024-10667-4

To recognize the potential for colon polyps to develop into cancer over time, early diagnosis is crucial for preventative healthcare. Timely identification significantly improves the prognosis and treatment outcomes for colorectal cancer patients. Image segmentation is crucial in medical image analysis for accurate diagnosis and treatment planning. Therefore, in this study, we present an alternative multilevel thresholding polyp segmentation method (MPOA) to enhance the segmentation of polyp images. The proposed method is based on enhancing the planet optimization algorithm (POA) by integrating operators from the reptile search algorithm (RSA). The evaluation of the developed MPOA is tested with different polyp images and compared with other image segmentation approaches. The results highlight the superior capability of MPOA, as evidenced by various performance measures in effectively segmenting polyp images. Furthermore, metrics such as peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), and fitness values demonstrate that MPOA outperforms the basic version of POA and other methods. The evaluation outcomes underscore the significant impact of RSA in enhancing the performance of POA for the segmentation of polyp images.