Classification of COVID-19 in X-ray images with Genetic Fine-tuning

• Published in: Computers & electrical engineering Vol. 96; p. 107467
• Main Authors: Vieira, Pablo A., Magalhães, Deborah M.V., Carvalho-Filho, Antonio O., Veras, Rodrigo M.S., Rabêlo, Ricardo A.L., Silva, Romuere R.V.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Ltd 01-12-2021; Elsevier BV
• Subjects: Convolutional neural networks; Coronaviruses; COVID-19; Evolutionary genetic systems; Fine-tuning; Image classification; Image processing; Machine learning; Medical imaging; Pneumonia; SARS-COV-2; Severe acute respiratory syndrome coronavirus 2; X-ray; Evolutionary genetic systems; Pneumonia; X-ray; Convolutional neural networks; SARS-COV-2; Fine-tuning
• ISSN: 0045-7906; 1879-0755
• DOI: 10.1016/j.compeleceng.2021.107467

New and more transmissible SARS-COV-2 variants aggravated the SARS-COV-2 emergence. Lung X-ray images stand out as an alternative to support case screening. The latest computer-aided diagnosis systems have been using Deep Learning (DL) to detect pulmonary diseases. In this context, our work investigates different types of pneumonia detection, including COVID-19, based on X-ray image processing and DL techniques. Our methodology comprehends a pre-processing step including data-augmentation, contrast enhancement, and resizing method to overcome the challenge of heterogeneous and few samples of public datasets. Additionally, we propose a new Genetic Fine-Tuning method to automatically define an optimal set of hyper-parameters of ResNet50 and VGG16 architectures. Our results are encouraging; we achieve an accuracy of 97% considering three classes: COVID-19, other pneumonia, and healthy. Thus, our methodology could assist in classifying COVID-19 pneumonia, which could reduce costs by making the process faster and more efficient.
[Display omitted]
•A methodology to automatically detect pneumonia caused by SARS-COV-2.•A Genetic Fine-Tuning applied to CNNs architectures for pneumonia classification.•Pre-processing step to overcome the scarcity and heterogeneity of pneumonia datasets.