Accurate detection of COVID-19 using deep features based on X-Ray images and feature selection methods

Abstract

COVID-19 is a severe epidemic affecting the whole world. This epidemic, which has a high mortality rate, affects the health systems and the economies of countries significantly. Therefore, ending the epidemic is one of the most important priorities of all states. For this, automatic diagnosis and detection systems are very important to control the epidemic. In addition to the recommendation of the “reverse transcription-polymerase chain reaction (RT-PCR)” test, additional diagnosis and detection systems are required. Hence, based on the fact that the COVID-19 virus attacks the lungs, automatic diagnosis and detection systems developed using X-ray and CT images come to the fore. In this study, a high-performance detection system was implemented with three different CNN (ResNet50, ResNet101, InceptionResNetV2) models and X-ray images of three different classes (COVID-19, Normal, Pneumonia). The particle swarm optimization (PSO) algorithm and ant colony algorithm (ACO) was applied among the feature selection methods, and their performances were compared. The results were obtained using support vector machines (SVM) and a k-nearest neighbor (k-NN) classifier using the 10-fold cross-validation method. The highest overall accuracy performance was 99.83% with the SVM algorithm without feature selection. The highest performance was achieved after the feature selection process with the SVM + PSO method as 99.86%. As a result, higher performance with less computational load has been achieved by realizing the feature selection. Based on the high results obtained, it is thought that this study will benefit radiologists as a decision support system.

Metaheuristics based COVID-19 detection using medical images: A review

“…The second most popular dataset is Dataset 5, which was used in six studies. The highest accuracy for this dataset is 99.86% [ 36 ]. CNN architecture was used along with PSO-ACO for feature selection.…”

“…Twelve studies [ 6 , 43 , 10 , 11 , 16 , 7 , 26 , 28 , 29 , 38 , 33 , 36 ], and [ 34 ] applied SVM for classification. These papers adopt metaheuristic algorithm MH-CovidNet, PSO-WOA, HSGO, CGRO, BDE, BD, DF, PSO, PSO, DE-PSO, PSO-ACO, and WO-IPSO for feature selection step, respectively.…”

“…For the last classification step, they prefer to use SVM. Among all [ 6 , 7 , 26 ], and [ 36 ] work with multi-class, while [ 10 , 11 , 28 , 29 , 31 , 33 , 34 , 43 ] deal with binary class classification problem. All studies report accuracy above 90%.…”

“…The experiments were performed on X-ray dataset 2 and the reported accuracy of this system was 99.81%. Narin and Ali [ 36 ] also used PSO, but it only worked with CNN. Meanwhile, PSO served as a feature selector, and SVM was trained for classification.…”

“…Albadr et al [ 20 ] reported a 100% accuracy for this dataset, but the experiments were conducted only on 200 images. Similarly, Dataset 5 has only 219 images [ 36 ]. The experiments performed on these datasets used deep learning models, which can produce biased results if only given limited data.…”