A Hybrid Method Based on LSTM and Optimized SVM for Diagnosis of Novel Coronavirus (COVID-19)

The coronavirus disease of 2019 (COVID-19), a unique Coronavirus strain, has created chaotic situation, negatively impacting the number deaths and people's lives globally. daily increase in COVID-19 instances is due to lack restricted availability detection techniques for determining disease's presence. Therefore, detecting positive results as soon feasible important preventing spread this epidemic treating infected people possible. As result these constraints, demand clinical decision-making systems based on predictive algorithms increased. article describes recurrent neural network (RNN) identifying (COVID-19) tries improve method. Different machine learning methodologies, such Support Vector Machines (SVMs), were used create system with deep algorithm called Long Short Term Memory (LSTM). research method tagged CT images patients. Various common picture features, central moments, Gabor wavelets, GLCM-related are discussed. Ant colony optimization-ant lion optimization (ACO-ALO) select optimum subsets SVM parameters. show that parameters penalty kernel have effect model correctness complexity. Besides, findings revealed proposed may be employed aid diagnose disease. uncover suggested strategy promising behavior terms increasing classification accuracies well optimal feature selection. Promisingly, presented can regarded useful tool clinicians.