Journal of Bionic Engineering ›› 2023, Vol. 20 ›› Issue (6): 2916-2934.doi: 10.1007/s42235-023-00392-4

• • 上一篇    下一篇

Illumination-Free Clustering Using Improved Slime Mould Algorithm for Acute Lymphoblastic Leukemia Image Segmentation

Krishna Gopal Dhal1; Swarnajit Ray2; Sudip Barik3; Arunita Das1   

  1. 1 Department of Computer Science and Application, Midnapore College (Autonomous), Paschim Medinipur, Midnapore, West Bengal, India  2 Department of Computer Science and Engineering, Maulana Abul Kalam Azad University of Technology, Kolkata, West Bengal, India  3 Department of Computer Science and Engineering, Kalyani Government Engineering College, Kalyani, West Bengal, India
  • 出版日期:2023-10-16 发布日期:2023-11-20
  • 通讯作者: Krishna Gopal Dhal E-mail:krishnagopal.dhal@midnaporecollege.ac.in
  • 作者简介:Krishna Gopal Dhal1; Swarnajit Ray2; Sudip Barik3; Arunita Das1

Illumination-Free Clustering Using Improved Slime Mould Algorithm for Acute Lymphoblastic Leukemia Image Segmentation

Krishna Gopal Dhal1; Swarnajit Ray2; Sudip Barik3; Arunita Das1   

  1. 1 Department of Computer Science and Application, Midnapore College (Autonomous), Paschim Medinipur, Midnapore, West Bengal, India  2 Department of Computer Science and Engineering, Maulana Abul Kalam Azad University of Technology, Kolkata, West Bengal, India  3 Department of Computer Science and Engineering, Kalyani Government Engineering College, Kalyani, West Bengal, India
  • Online:2023-10-16 Published:2023-11-20
  • Contact: Krishna Gopal Dhal E-mail:krishnagopal.dhal@midnaporecollege.ac.in
  • About author:Krishna Gopal Dhal1; Swarnajit Ray2; Sudip Barik3; Arunita Das1

摘要: Partitional clustering techniques such as K-Means (KM), Fuzzy C-Means (FCM), and Rough K-Means (RKM) are very simple and efective techniques for image segmentation. But, because their initial cluster centers are randomly determined, it is often seen that certain clusters converge to local optima. In addition to that, pathology image segmentation is also problematic due to uneven lighting, stain, and camera settings during the microscopic image capturing process. Therefore, this study proposes an Improved Slime Mould Algorithm (ISMA) based on opposition based learning and diferential evolution’s mutation strategy to perform illumination-free White Blood Cell (WBC) segmentation. The ISMA helps to overcome the local optima trapping problem of the partitional clustering techniques to some extent. This paper also performs a depth analysis by considering only color components of many well-known color spaces for clustering to fnd the efect of illumination over color pathology image clustering. Numerical and visual results encourage the utilization of illumination-free or color component-based clustering approaches for image segmentation. ISMA-KM and “ab” color channels of CIELab color space provide best results with above-99% accuracy for only nucleus segmentation. Whereas, for entire WBC segmentation, ISMA-KM and the “CbCr” color component of YCbCr color space provide the best results with an accuracy of above 99%. Furthermore, ISMA-KM and ISMA-RKM have the lowest and highest execution times, respectively. On the other hand, ISMA provides competitive outcomes over CEC2019 benchmark test functions compared to recent well-established and efcient Nature-Inspired Optimization Algorithms (NIOAs).

关键词: Pathology image , · Image segmentation , · Clustering , · Color space , · White blood cell , · Optimization , · Swarm intelligence , · Fuzzy clustering , · Rough clustering

Abstract: Partitional clustering techniques such as K-Means (KM), Fuzzy C-Means (FCM), and Rough K-Means (RKM) are very simple and efective techniques for image segmentation. But, because their initial cluster centers are randomly determined, it is often seen that certain clusters converge to local optima. In addition to that, pathology image segmentation is also problematic due to uneven lighting, stain, and camera settings during the microscopic image capturing process. Therefore, this study proposes an Improved Slime Mould Algorithm (ISMA) based on opposition based learning and diferential evolution’s mutation strategy to perform illumination-free White Blood Cell (WBC) segmentation. The ISMA helps to overcome the local optima trapping problem of the partitional clustering techniques to some extent. This paper also performs a depth analysis by considering only color components of many well-known color spaces for clustering to fnd the efect of illumination over color pathology image clustering. Numerical and visual results encourage the utilization of illumination-free or color component-based clustering approaches for image segmentation. ISMA-KM and “ab” color channels of CIELab color space provide best results with above-99% accuracy for only nucleus segmentation. Whereas, for entire WBC segmentation, ISMA-KM and the “CbCr” color component of YCbCr color space provide the best results with an accuracy of above 99%. Furthermore, ISMA-KM and ISMA-RKM have the lowest and highest execution times, respectively. On the other hand, ISMA provides competitive outcomes over CEC2019 benchmark test functions compared to recent well-established and efcient Nature-Inspired Optimization Algorithms (NIOAs).

Key words: Pathology image , · Image segmentation , · Clustering , · Color space , · White blood cell , · Optimization , · Swarm intelligence , · Fuzzy clustering , · Rough clustering