Dynamic colormap visualization integrated with Harris hawks optimization for enhanced lung CT segmentation and diagnostic precision

• Published in: Cluster computing Vol. 28; no. 6; p. 377
• Main Authors: Drogham, Osama, Ryalat, Mohammad H., Al-Najdawi, Nijad, Alkhawaldeh, Rami S., AlShaqsi, Jamil, Al-Betar, Mohammed Azmi
• Format: Journal Article
• Language: English
• Published: New York Springer US 01.10.2025; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Chronic obstructive pulmonary disease; Computed tomography; Computer Communication Networks; Computer Science; COVID-19; Data integrity; Deep learning; Illnesses; Image segmentation; Lung diseases; Medical imaging; Medical prognosis; Neural networks; Operating Systems; Optimization; Optimization techniques; Pneumonia; Processor Architectures; Respiratory diseases; Visualization; Dynamic colormap visualization; Multi-level thresholding; Harris hawks optimization (HHO); Medical imaging; Lung CT segmentation
• ISSN: 1386-7857; 1573-7543
• DOI: 10.1007/s10586-025-05220-4

This study presents a novel method that utilizes Harris Hawks Optimization combined with dynamic colormap visualization to enhance the quality of lung CT scan segmentation. The Harris hawks optimization algorithm is a swarm-based method used to enhance multi-level thresholding for image segmentation, hence facilitating the identification of regions of interest (ROIs) in medical images. An analysis of different colormap schemes including Accent, Gray, Hot, Inferno and Jet, was conducted to improve the visualization of segmented images. The experimental results show the efficiency of the HHO algorithm from the segmentation accuracy perspective as compared to the conventional optimization techniques using publicly available datasets from the Cancer Imaging Archive. In particular, the average SSIM was above 98% while the Jaccard Index was more than 90%. The expert evaluation confirms earlier findings that using the HHO algorithm with the Inferno colormap, particularly with four or five thresholds, achieves optimal image clarity and diagnostic value for clinical purposes. In addition, the method provides a promising way to enhance diagnostic precision and treatment strategies for lung diseases, making it highly valuable for pulmonary healthcare, particularly in urgent scenarios such as pandemics.