Auto-Diagnosis of COVID-19 Using Lung CT Images With Semi-Supervised Shallow Learning Network

In the current world pandemic situation, contagious Novel Coronavirus Disease 2019 (COVID-19) has raised a real threat to human lives owing infection on lung cells and respiratory systems. It is daunting task for researchers find suitable patterns CT images automated diagnosis of COVID-19. A novel integrated semi-supervised shallow neural network framework comprising Parallel Quantum-Inspired Self-supervised Network (PQIS-Net) automatic segmentation followed by Fully Connected (FC) layers, proposed in this article. The PQIS-Net model aimed at providing fully slices without incorporating pre-trained convolutional based models. parallel trinity layered structure quantum bits are interconnected using an N-connected second order neighborhood-based topology suggested architecture with wide variations local intensities. random patch-based classification segmented incorporated layers framework. Intensive experiments have been conducted three publicly available data sets, one purely other two (COVID-19 diagnosis). experimental outcome self-supervised efficiency (Accuracy, Precision AUC) found be promising. also superior than best state-of-the-art techniques network-based models, specially COVID-19 Mycoplasma Pneumonia (MP) screening.