Detailed Explanation: Optimizing COVID-19 CT-Scan Classification with Feature Engineering and Genetic Algorithm

This research delves into the optimization of a genetic algorithm for the categorization of COVID-19 CT scan images through the amalgamation of a high-level grayscale cooccurrence matrix (GCLM) elimination technique, color consistency vector (CCV), and discrete wavelet transform (DWT). The primary objective of this combination is to enhance the discriminatory capabilities of specific criteria, thereby elevating the precision of the machine learning model. Specifically, the crafting of the classification was executed through Fisher-KNN (F-KNN), linear support vector machine (L-SVM), and the Fisher visual decision tree (F-Tree).Genetic algorithms were employed for targeted selection to amplify the efficiency of these models in accurately identifying COVID-19 samples within CT images. The outcomes of this investigation offer valuable insights for continuous endeavors aimed at enhancing COVID-19 diagnostics by integrating architecture, genetic algorithms, and machine learning models. Notably, among the proposed methodologies, the fuzzy tree classifier achieved an impressive accuracy of 93.55%, while alternative classifiers such as Naive Bayes, L-SVM, and F-KNN attained accuracies of 92.6%, 92.12%, and 92.33% respectively. This research underscores the significance of leveraging advanced computational techniques in the field of medical imaging for improved disease detection and classification. The fusion of genetic algorithms with machine learning frameworks unveils an exciting pathway to boost the precision and effectiveness of COVID-19 detection via CT imaging. The results of this research play a significant role in the evolving conversation surrounding the incorporation of cutting-edge technologies in the medical field, aiming for enhanced disease control and accurate diagnosis.