Fine-Tuning Convolutional Neural Networks for COVID-19 Detection from Chest X-ray Images

Abstract

As the COVID-19 pandemic continues to ravage the world, the use of chest X-ray (CXR) images as a complementary screening strategy to reverse transcription-polymerase chain reaction (RT-PCR) testing continues to grow owing to its routine clinical application to respiratory diseases. We performed extensive convolutional neural network (CNN) fine-tuning experiments and identified that models pretrained on larger out-of-domain datasets show an improved performance. This suggests that a priori knowledge of models from out-of-field training should also apply to X-ray images. With appropriate hyperparameters selection, we found that higher resolution images carry more clinical information, and the use of mixup in training improved the performance of the model. The experimental showed that our proposed transfer learning present state-of-the-art results. Furthermore, we evaluated the performance of our model with a small amount of downstream training data and found that the model still performed well in COVID-19 identification. We also explored the mechanism of model detection using a gradient-weighted class activation mapping (Grad-CAM) method for CXR imaging to interpret the detection of radiology images. The results helped us understand how the model detects COVID-19, which can be used to discover new visual features and assist radiologists in screening.

A Review Study of the Deep Learning Techniques used for the Classification of Chest Radiological Images for COVID-19 Diagnosis

“…The use of transfer learning was still very common in 2021. Some scientists concentrated on the power of transfer learning ( ( Chakraborty et al, 2021 ) ( Fayemiwo et al, 2021 ) ( Gupta et al, 2022 ) ( Zhao et al, 2021 )). They employed various classical Deep Learning architectures with transfer learning and make a comparison about the networks’ performance.…”

“… Parka et al ( Park et al, 2021 ) Apr-21 Chest X-Ray Valencian Region Medical Image Bank [BIMCV] (De La Iglesia Vay´a et al, 2020), Brixia (Signoroni et al, 2020b) National Institutes of Health [NIH] (Wang et al, 2017) CheXpert He et al ( He et al, 2021 ) May-21 Chest CT COVID-19 CT Lung and Infection Segmentation Dataset, v1.0. (M.Jun, G.Cheng, 2020) MosMedData: Chest CT Scans with COVID-19 Related Findings, 2020 COVID-19 radiology-data collection and preparation for Artificial Intelligence (H. B. Jenssen, 2020) Sangeetha et al ( Sangeetha et al, 2021 ) Aug-21 Chest CT Radiological Society of North America (RSNA) Zhao et al ( Zhao et al, 2021 ) Aug-21 Chest X-Ray The RSNA International COVID-19 Open Radiology Database (RICORD). (Tsai, E.B.…”

COVID-19 Detection on Chest X-Ray Images: A comparison of CNN architectures and ensembles

“…Pavlova et al (2021) proposed the COVIDx8B dataset, the largest and most diverse COVID-19 CXR dataset in open access form, and the COVID-Net CXR-2 model, a CNN specially tailored for COVID-19 detection on CXR images using machine-driven design, which achieved an accuracy of 95.5%. Zhao et al (2021) used ResNet50V2 to classify the COVIDx8B dataset with an accuracy of 96.5% in the best scenario. Dominik (2021) proposed a lightweight architecture called BaseNet and achieved an accuracy of 95.50% on COVIDx8B.…”

“…However, most of them used relatively small and more homogeneous datasets. The largest and most diverse COVID-19 CXR benchmark dataset available so far is COVIDx8B 1 (Zhao et al, 2021). It has 16, 352 CXR images, from which 2, 358 are COVID-19 positive and the remaining are from both healthy and pneumonia patients.…”

“…Therefore it is larger and more heterogeneous than earlier available datasets. However, there are only a few works that used this dataset so far (Pavlova et al, 2021;Zhao et al, 2021;Dominik, 2021). A recent survey on applications of artificial intelligence in COVID-19 pandemic (Khan et al, 2021) reviewed dozens of papers, including 16 papers on CNNs applied to CXR images and all of them used earlier available datasets which are smaller than COVIDx8B.…”