Lung infection detection and classification using the integration of the improved grasshopper and the remora optimization approaches with improved SVM

• Published in: Neural computing & applications Vol. 37; no. 27; pp. 22573 - 22591
• Main Author: Bhimavarapu, Usharani
• Format: Journal Article
• Language: English
• Published: London Springer London 01.09.2025
• Subjects: Artificial Intelligence; Computational Biology/Bioinformatics; Computational Science and Engineering; Computer Science; Data Mining and Knowledge Discovery; Image Processing and Computer Vision; Probability and Statistics in Computer Science; S.I.: Hybrid Approaches to Nature-inspired Optimization Algorithms and Their Applications; Special Issue on Hybrid Approaches to Nature-inspired Optimization Algorithms and Their Applications; Grasshopper optimization; Lung disease; Rider optimization; Multi-thresholding
• ISSN: 0941-0643; 1433-3058
• DOI: 10.1007/s00521-024-10624-1

Infectious lung diseases, such as pneumonia and COVID-19, pose significant threats to global health, with high mortality rates and substantial burdens on healthcare systems. Accurate and timely diagnosis is crucial for effective management and treatment. This study addresses the limitations of existing diagnostic methods by proposing advanced techniques based on computer-aided diagnosis systems and enhanced machine-learning algorithms. The methodology involves the development of novel algorithms for image enhancement, segmentation, feature selection, and classification. A kurtosis-based multi-thresholding grasshopper optimization algorithm is proposed for image segmentation, reducing complexity and enhancing the accuracy of lesion identification. An improved rider optimization algorithm is also introduced for feature selection, aiming to prioritize relevant features and reduce dimensionality effectively. Furthermore, an enhanced support vector machine (SVM) algorithm for lesion classification is presented, utilizing linear mapping to generate feature scores for regions of interest. This facilitates the evaluation of the loss function and improves classification results. The approach’s effectiveness is demonstrated using datasets comprising chest X-ray and CT scan images from the LIDC-IDRI and Montgomery datasets. The improved optimization algorithms were trained and tested over the chest X-ray and CT scan image datasets. An improved SVM classified the lesions with an accuracy of 99.9% for chest X-ray images and 99.8% for CT scan images. The results proved that the improved SVM adequately classifies lung diseases from the chest X-ray and CT scan images. The findings suggest that the proposed methodologies significantly enhance the accuracy and efficiency of diagnosing pneumonia and COVID-19 from medical images. By addressing the limitations of existing diagnostic techniques, this research contributes to improving healthcare practices and ultimately reducing the burden of infectious lung diseases on a global scale.