An efficient multilevel thresholding image segmentation method based on the slime mould algorithm with bee foraging mechanism: A real case with lupus nephritis images. (March 2022)

• Record Type: Journal Article
• Authors: Chen, Xiaowei; Huang, Hui; Heidari, Ali Asghar; Sun, Chuanyin; Lv, Yinqiu; Gui, Wenyong; Liang, Guoxi; Gu, Zhiyang; Chen, Huiling; Li, Chengye; Chen, Peirong
• Is Part Of: Computers in biology and medicine. Volume 142(2022)
• Publication Date: 2022-03
• Subjects: Slime mould algorithm -- Diagnosis -- Image -- Meta-heuristic -- Swarm-intelligence -- Lupus nephritis -- Artificial bee colony -- Multilevel thresholding image segmentation; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2021.105179
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: To improve the diagnosis of Lupus Nephritis (LN), a multilevel LN image segmentation method is developed in this paper based on an improved slime mould algorithm. The search of the optimal threshold set is key to multilevel thresholding image segmentation (MLTIS). It is well known that swarm-based methods are more efficient than the traditional methods because of the high complexity in finding the optimal threshold, especially when performing image partitioning at high threshold levels. However, swarm-based methods tend to obtain the poor quality of the found segmentation thresholds and fall into local optima during the process of segmentation. Therefore, this paper proposes an ASMA-based MLTIS approach by combining an improved slime mould algorithm (ASMA), where ASMA is mainly implemented by introducing the position update mechanism of the artificial bee colony (ABC) into the SMA. To prove the superiority of the ASMA-based MLTIS method, we first conducted a comparison experiment between ASMA and 11 peers using 30 test functions. The experimental results fully demonstrate that ASMA can obtain high-quality solutions and almost does not suffer from premature convergence. Moreover, using standard images and LN images, we compared the ASMA-based MLTIS method with other peers and evaluated the segmentation results using three evaluation indicators called PSNR, SSIM, and FSIM. The proposed ASMA can be an excellent swarm intelligence optimization method that can maintain a delicate balance during the segmentation process of LN images, and thus the ASMA-based MLTIS method has great potential to be used as an image segmentation method for LN images. The lastest updates for the SMA algorithm are available in https://aliasgharheidari.com/SMA.html . Highlights: This paper proposes an improved slime mould algorithm namely ASMA. ASMA is enhanced by introducing the bee foraging mechanism to SMA. An effective lupus nephritis image segmentation method based on ASMA is developed. The segmentation method is evaluated under some representative threshold levels. ASMA extremely improves the solution and image segmentation quality.