ABSTRACT
Purpose: In early 2020, the world is amid a significant pandemic due to the novel coronavirus disease outbreak, commonly called the COVID-19. Coronavirus is a lung infection disease caused by the Severe Acute Respiratory Syndrome Coronavirus 2 virus (SARS-CoV-2). Because of its high transmission rate, it is crucial to detect cases as soon as possible to effectively control the spread of this pandemic and treat patients in the early stages. RT-PCR-based kits are the current standard kits used for COVID-19 diagnosis, but these tests take much time despite their high precision. A faster automated diagnostic tool is required for the effective screening of COVID-19. Methods: In this study, a new semi-supervised feature learning technique is proposed to screen COVID-19 patients using chest CT scans. The model proposed in this study uses a three-step architecture, consisting of a convolutional autoencoder based unsupervised feature extractor, a multi-objective genetic algorithm (MOGA) based feature selector, and a Bagging Ensemble of support vector machines based binary classifier. The proposed architecture has been designed to provide precise and robust diagnostics for binary classification (COVID vs.nonCOVID). A dataset of 1252 COVID-19 CT scan images, collected from 60 patients, has been used to train and evaluate the model. Results: The best performing classifier within 127 ms per image achieved an accuracy of 98.79%, the precision of 98.47%, area under curve of 0.998, and an F1 score of 98.85% on 497 test images. The proposed model outperforms the current state of the art COVID-19 diagnostic techniques in terms of speed and accuracy. Conclusion: The experimental results prove the superiority of the proposed methodology in comparison to existing methods.The study also comprehensively compares various feature selection techniques and highlights the importance of feature selection in medical image data problems.
ABSTRACT
In the current world pandemic situation, the contagious Novel Coronavirus Disease 2019 (COVID-19) has raised a real threat to human lives owing to infection on lung cells and human respiratory systems. It is a daunting task for the researchers to find suitable infection patterns on lung CT images for automated diagnosis of COVID-19. A novel integrated semi-supervised shallow neural network framework comprising a Parallel Quantum-Inspired Self-supervised Network (PQIS-Net) for automatic segmentation of lung CT images followed by Fully Connected (FC) layers, is proposed in this article. The proposed PQIS-Net model is aimed at providing fully automated segmentation of lung CT slices without incorporating pre-trained convolutional neural network based models. A parallel trinity of layered structure of quantum bits are interconnected using an N-connected second order neighborhood-based topology in the suggested PQIS-Net architecture for segmentation of lung CT slices with wide variations of local intensities. A random patch-based classification on PQIS-Net segmented slices is incorporated at the classification layers of the suggested semi-supervised shallow neural network framework. Intensive experiments have been conducted using three publicly available data sets, one for purely segmentation task and the other two for classification (COVID-19 diagnosis). The experimental outcome on segmentation of CT slices using self-supervised PQIS-Net and the diagnosis efficiency (Accuracy, Precision and AUC) of the integrated semi-supervised shallow framework is found to be promising. The proposed model is also found to be superior than the best state-of-the-art techniques and pre-trained convolutional neural network-based models, specially in COVID-19 and Mycoplasma Pneumonia (MP) screening. CCBY