Registration of COVID-19-Affected Lung CT Images Using Teaching-Learning-Based Optimization (TLBO)
摘要
Meta-heuristic algorithms are extensively used in medical image registration to address optimization challenges and aid in diagnosing various illnesses, including cancers such as breast, lung, and brain cancer. Leveraging image transformation, intensity-based registration techniques in both multi-modal and mono-modal registration combine several images containing identical information into a single representation. Enhancing the similarity metric across different images is crucial. In this study, we propose a novel approach for detecting COVID-19-impacted lung regions in Computed Tomography (CT) images by addressing image registration. After evaluating the success of the Teaching Learning Based Optimization (TLBO) algorithm in medical image analysis and comparisons, we propose a TLBO algorithm for registering lung CT images of COVID-19 patients. Simulation results demonstrate that the proposed algorithm surpasses the Particle Swarm Optimization (PSO) algorithm in both registration accuracy and robustness.