Optimized chest X-ray image semantic segmentation networks for COVID-19 early detection

Abstract

BACKGROUND: Although detection of COVID-19 from chest X-ray radiography (CXR) images is faster that PCR sputum testing, the accuracy of detecting COVID-19 from CXR images is lacking in the existing deep learning models. OBJECTIVE: This study aims to classify COVID-19 and normal patients from CXR images using semantic segmentation networks for detecting and labeling COVID-19 infected lung lobes in CXR images. METHODS: For semantically segmenting infected lung lobes in CXR images for COVID-19 early detection, three structurally different deep learning (DL) networks such as SegNet, U-Net and hybrid CNN with SegNet plus U-Net, are proposed and investigated. Further, the optimized CXR image semantic segmentation networks such as GWO SegNet, GWO U-Net, and GWO hybrid CNN are developed with the grey wolf optimization (GWO) algorithm. The proposed DL networks are trained, tested, and validated without and with optimization on the openly available dataset that contains 2,572 COVID-19 CXR images including 2,174 training images and 398 testing images. The DL networks and their GWO optimized networks are also compared with other state-of-the-art models used to detect COVID-19 CXR images. RESULTS: All optimized CXR image semantic segmentation networks for COVID-19 image detection developed in this study achieved detection accuracy higher than 92%. The result shows the superiority of optimized SegNet in segmenting COVID-19 infected lung lobes and classifying with an accuracy of 98.08% compared to optimized U-Net and hybrid CNN. CONCLUSION: The optimized DL networks has potential to be utilised to more objectively and accurately identify COVID-19 disease using semantic segmentation of COVID-19 CXR images of the lungs.

A novel metaheuristic prediction approach for COVID-19 cases using XGBoost algorithm

“…Weights are adjustable depending on the metric chosen [26].This method works well for data such as variable as Covid-19 data, but another crucial part is dividing the data for training.We need to have enough historical data for the model to train well which results in more accurate prediction or test trials.Part of the data is classified as training data and the rest is classified as testing, with a 70:30 training validation split.Here is where we need to look at some of the most relevant parameters of the model which are: learning rate, number of leaves, and the maximum depth. The loss function graph and the number of iterations are being observed when adjusting these parameters [27]. The type of dataset is based on time series and is summed by the total number of cases per day.…”

“…For the data analysis, it is inevitable to check whether the data is seasonal or not and whether there is any autocorrelation, in order to get a better sense of choosing the parameters for the model [24].In [25][26][27][28][29] provided the segmentation of the CXR images in diagnosis of COVID-19 in Chest X-ray images. As seen in [Figure 3] the data was quite seasonal from October to July, and the it lowers, and it goes back up in the following October.…”

Deep learning framework for classification of COVID-19 from Chest X-rays

“…In order to aid radiologists and detect the Covid-19 virus, the goal of this study is to present a framework connecting training and testing that makes use of deep CNN classifiers. In [20], authors implemented segmentation using UNet and SegNet as preprocessing to detect COVID-19 using chest X-ray images. The authors of [21] conducted experiments to determine whether or not CT scan pictures could accurately identify viruses.…”

Directional mutation and crossover boosted ant colony optimization with application to COVID-19 X-ray image segmentation

“…Today, many COVID-19 X-ray image segmentation methods have been proposed by researchers. Gopatoti et al [ 43 ] presented a convolutional neural network chest X-ray radiography image segmentation model that combined SegNet, U-Net, and the gray wolf optimization algorithm. Tahir et al [ 44 ] proposed a novel method with segmentation networks, U-Net++, and feature pyramid networks for COVID-19 lung image segmentation.…”