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

Pablo A Vieira; Deborah M V Magalhães; Antonio O Carvalho-Filho; Rodrigo M S Veras; Ricardo A L Rabêlo; Romuere R V Silva

doi:10.1016/j.compeleceng.2021.107467

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

Comput Electr Eng. 2021 Dec:96:107467. doi: 10.1016/j.compeleceng.2021.107467. Epub 2021 Sep 24.

Authors

Pablo A Vieira¹, Deborah M V Magalhães^{1

2}, Antonio O Carvalho-Filho^{1

2}, Rodrigo M S Veras³, Ricardo A L Rabêlo^{1

3}, Romuere R V Silva^{1

2

3}

Affiliations

¹ Electrical Engineering, Federal University of Piauí, Teresina, Brazil.
² Information Systems, Federal University of Piauí, Picos, Brazil.
³ Computer Science, Federal University of Piauí, Picos, Brazil.

Abstract

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.

Keywords: Convolutional neural networks; Evolutionary genetic systems; Fine-tuning; Pneumonia; SARS-COV-2; X-ray.