Artificial Neural Network-based COVID-19 Diagnosis and Prediction

This thesis focuses on developing extreme learning machine-based learning algorithms and using them to train a convolutional neural network model to detect COVID-19 from chest X-ray image data and optimise the hyperparameters of a long short-term memory-based model to forecast the number of daily new cases infected with COVID-19. The models show excellent performance on both COVID-19 detection and daily new cases prediction. Thus, this thesis will be of great reference if some other pandemics occur in the future.