Registration of Lung CT-Images Affected with Covid-19 Using Centroid Opposition-Based Multiverse Optimizer

• Published in: SN computer science Vol. 6; no. 3; p. 196
• Main Authors: Nayak, Somen, Sarkar, Achyuth, Mondal, Subhodip
• Format: Journal Article
• Language: English
• Published: Singapore Springer Nature Singapore 19.02.2025; Springer Nature B.V
• Subjects: Advances on Human Centered Computing for Society 5.0; Algorithms; Centroids; Computed tomography; Computer Imaging; Computer Science; Computer Systems Organization and Communication Networks; COVID-19; Data Structures and Information Theory; Image manipulation; Image quality; Image registration; Information Systems and Communication Service; Lungs; Medical imaging; Medical research; Methods; Optimization techniques; Original Research; Particle swarm optimization; Pattern Recognition and Graphics; Registration; Search algorithms; Similarity measures; Software Engineering/Programming and Operating Systems; Statistical methods; Viruses; Vision; Multi modality; Feature-based registration; Meta-heuristics algorithm; Mono modality image registration; Medical image; Intensity-based registration; Similarity metric; Mutual information
• ISSN: 2661-8907; 2662-995X; 2661-8907
• DOI: 10.1007/s42979-024-03654-y

Medical image registration can be used to address a variety of practical difficulties that are related to the collection of medical pictures from multiple modalities. Multimodal registration uses intensity-based registration approaches, and these strategies will amalgamate supplementary images that possess the same content into a singular representation through image manipulation. It is crucial to optimize the modification of computed tomography (CT) scans. The improvement of the similarity metric between the various scans is really necessary.Numerous optimization methods have been proposed recently, focusing on the foundational elements of optimization. However, from the perspectives of efficiency and quality, there is still a lot that may be improved. For diagnosis, Lung CT image registration is crucial in identifying the spot differences between the various Lung CT images. The registration of lung CT scans using multiverse optimization (MVO) with computing opposition-based learning followed by chaotic-based initialization is suggested in this research. We have proposed the Meta heuristics-based Optimization (MVO_COBL) algorithm to register lung CT images for both multimodal and monomodal medical image registration. We first collected CT scans of the lungs from the data set.Images are then processed using MVO_COBL, and the registered images are finally extracted. The suggested lung CT image registration method’s outcomes are contrasted with those of the hybrid particle swarm optimization with Grey wolf optimizer (HPSGWO), crow search algorithms (CSA), Grey wolf optimizer (GWO), and particle swarm optimization (PSO) based methods. The experimental results statistically speaking that, the suggested method performs better than HPSGWO, PSO, GWO and CSA methods in registering lung CT images.