Feature Based Automated Detection of COVID-19 from Chest X-Ray Images

Abstract

Nowadays the biggest challenge for health care is controlling the pandemic of Coronavirus disease 2019 . Radiological investigation combining with machine learning can serve as a standardized methodology for detecting COVID-19. Chest X-ray imaging is the most feasible radiological test for COVID-19. Machine learning-based automated classification of COVID-19 from chest X-ray images can act as an assistive method to the medical experts for accurate diagnosis of disease. Aiming at this, the study focused on developing a simplified method of X-ray image based computerized COVID-19 detection through conventional feature extraction and classification approach. The method of X-ray image based COVID-19 detection consists of only two main steps: feature extraction, and classification. In feature extraction, a total of 55 X-ray image texture features is extracted from seven different groups. Classification of COVID-19 has been performed using those extracted features through four different popularly used classifiers. The overall analysis of the study has been performed over two datasets. The Random Forest classifier generates the best accuracy of 98.6% and 98.9% for dataset 1 and 2 with the area under the curve (AUC) values 0.99 and 1 respectively. The outcome of our study provides optimal accuracy of COVID-19 classification using X-ray images compare to existing popular studies in this domain. The reliable and less-complex feature of the proposed method may serve it as a computerized X-ray image based COVID-19 detection mechanism, especially in rural areas where medical experts are not available.

Rapid and Accurate Diagnosis of COVID-19 Cases from Chest X-ray Images through an Optimized Features Extraction Approach

“…Dissimilar geometry structures, shapes, and sizes are present in each COVID-19 infected image for different levels of infections. From the previous works [29][30][31][32], we understand that the shape descriptors such as Zernike and Hu moments [24] are good enough to define detailed shape information, especially for medical images (CT images or X-ray images). A good shape descriptor should provide the following: High discrimination ability with low redundancy; Be invariant to scale, rotation, and translation; and Provide both coarse to finer detailed representation.…”

“…However, when, T = 0.15, the reported sensitivity and specificity were 96% and 70.6%, respectively. COVID-19 infection detection through the feature extraction method and classification model is discussed in [24], Bardhan et al, 2021. In feature extraction, the total number of 55 texture features are gathered from Xray images used for training and testing. The features are classified in the COVID-19 region using four standard classification models.…”

Desenvolvimento de Método de Acesso Métrico com Calibração de Distância Perceptual para Consultas por Similaridade de Imagens Médicas

“…No contexto de imagens pulmonares relacionadas ao COVID-19 e de imagens de ultrassom para identificação de tumores em mamas, as características de textura têm sido amplamente utilizadas (BENAOUALI et al, 2022;REZAZADEH;JAFARIAN;KORD, 2022;BARDHAN;ROGA, 2021;PEREIRA et al, 2020). Desta forma, em ambas as bases de imagens adquiridas, foram extraídas características de nove extratores de textura e um extrator de histograma de cinzas que foram abordados na Seção 2.3 e utilizados para experimentos do método desenvolvido nesta pesquisa.…”