On the Use of Deep Learning for Imaging-Based COVID-19 Detection Using Chest X-rays

Abstract

The global COVID-19 pandemic that started in 2019 and created major disruptions around the world demonstrated the imperative need for quick, inexpensive, accessible and reliable diagnostic methods that would allow the detection of infected individuals with minimal resources. Radiography, and more specifically, chest radiography, is a relatively inexpensive medical imaging modality that can potentially offer a solution for the diagnosis of COVID-19 cases. In this work, we examined eleven deep convolutional neural network architectures for the task of classifying chest X-ray images as belonging to healthy individuals, individuals with COVID-19 or individuals with viral pneumonia. All the examined networks are established architectures that have been proven to be efficient in image classification tasks, and we evaluated three different adjustments to modify the architectures for the task at hand by expanding them with additional layers. The proposed approaches were evaluated for all the examined architectures on a dataset with real chest X-ray images, reaching the highest classification accuracy of 98.04% and the highest F1-score of 98.22% for the best-performing setting.

A Novel CovidDetNet Deep Learning Model for Effective COVID-19 Infection Detection Using Chest Radiograph Images

“…The comparison results are shown in Table 11, elaborating the success of the proposed CovidDetNet model in identifying COVID-19 from chest radiograph images compared to existing alternatives. It is important to mention that the proposed model outperformed the approaches [43,44] using the same dataset (COVID-19 radiography database) for COVID-19 detection. It is worth noting that these approaches are more computationally challenging than the proposed approach because they use deeper models, leading to overfitting.…”

“…The COVID-19 images dataset has also been improved to improve its performance. Additionally, Okolo et al [44] employed eleven CNN models to classify chest radiograph images as belonging to healthy persons, people with COVID-19, or people with viral pneumonia. They analyzed three distinct improvements to modify the frameworks for the COVID-19 detection by expanding them with extra layers.…”

“…This experiment aims to recover the performance of the proposed CovidDetNet framework in identifying and classifying COVID-19 from chest radiograph images compared to existing state-of-the-art deep learning approaches in the literature. We compared the proposed classification model performance to various approaches [43,44,68,69]. Prateek et al [68] proposed an automated diagnostic method for COVID-19 detection and classification using the DL model on chest radiograph images.…”

“…The proposed techniques were tested and analyzed on a chest X-ray images dataset, with the best-performing setup achieving the maximum classification accuracy of 98.04% and the highest F1-score of 98.22%. Azhar et al [44] introduced a CNN framework for detecting COVID-19 using chest radiographs that are faster and more reliable. For feature extraction, a CNN approach was utilized.…”

How Is Mass Spectrometry Tackling the COVID-19 Pandemic?

“…These profiles can be the basis for developing clinical testing for COVID-19 severity prognosis, which relies on targeted strategies using reliable and low-cost effective (accessible) instrumentation. Examples of such diagnostic methods that can be used to determine the severity of COVID-19 patients are image-based diagnostics looking for lung inflammation (Docherty et al, 2020;Guan et al, 2020;Liu et al, 2020;Okolo et al, 2021) and blood/urine-based approaches looking for inflammation, cardiovascular, diabetes-type, and liver dysfunction biomarkers (Gross et al, 2020;Gross et al, 2021;Siemens Healthineers. Lev, 2021).…”