Classification of COVID-19 chest X-Ray and CT images using a type of dynamic CNN modification method

Abstract

Understanding and classifying Chest X-Ray (CXR) and computerised tomography (CT) images are of great significance for COVID-19 diagnosis. The existing research on the classification for COVID-19 cases faces the challenges of data imbalance, insufficient generalisability, the lack of comparative study, etc. To address these problems, this paper proposes a type of modified MobileNet to classify COVID-19 CXR images and a modified ResNet architecture for CT image classification. In particular, a modification method of convolutional neural networks (CNN) is designed to solve the gradient vanishing problem and improve the classification performance through dynamically combining features in different layers of a CNN. The modified MobileNet is applied to the classification of COVID-19, Tuberculosis, viral pneumonia (with the exception of COVID-19), bacterial pneumonia and normal controls using CXR images. Also, the proposed modified ResNet is used for the classification of COVID-19, non-COVID-19 infections and normal controls using CT images. The results show that the proposed methods achieve 99.6% test accuracy on the five-category CXR image dataset and 99.3% test accuracy on the CT image dataset. Six advanced CNN architectures and two specific COVID-19 detection models, i.e., COVID-Net and COVIDNet-CT are used in comparative studies. Two benchmark datasets and a CXR image dataset which combines eight different CXR image sources are employed to evaluate the performance of the above models. The results show that the proposed methods outperform the comparative models in classification accuracy, sensitivity, and precision, which demonstrate their potential in computer-aided diagnosis for healthcare applications.

Keywords: COVID-19 detection; Chest X-Ray and CT images; Deep learning; Modified CNN.

Fig. 1

Diagram of the implementation process.

Fig. 2

Example CXR images from data sources [24,[36], [37], [38], [39]].

1. •

   COVID-19 CXR images are from the open source GitHub repository ieee8023/covid-chestxray-dataset (DS 1) [36], Actualmed-COVID-chestxray-dataset (DS 2), Fig. 1-COVID-chestx-ray-dataset (DS 3) [37], COVID-19 Radiography Database (DS 4) [24]. Data preparing for COVID-19 images in this paper referred to the code provided in the GitHub repository COVIDx Dataset contributed by Linda Wang et al. [11].

2. •

   The images of tuberculosis positive cases are from the dataset [38] which include CXR databases respectively obtained in Shenzhen, China and Montgomery, USA (DS 5), and the data source TB Portals Program, Office of Cyber Infrastructure and Computational Biology (OCICB), National Institute of Allergy and Infectious Diseases (NIAID) (DS 6).

3. •

   The bacterial and viral pneumonia CXR images are from Pneumonia Classification Dataset (DS 7) [39].

4. •

   The CXR images of normal controls are from COVID-19 Radiography Database (DS 4) [24].

Fig. 3

Validation accuracy for six CNN models on the Chest X-Ray image dataset.

Fig. 4

The structure of the Modified MobileNet. The contents of the original MobileNetv3_small are kept and denoted by light blue, whereas the dark blue blocks represent the modified parts. $w_1,\dots ,w_5$ are the weights respectively multiplied by the outputs of each pointwise conv block. $\times$ represents element-wise multiplication, $\oplus$ denotes the addition of five outputs of pointwise conv blocks multiplied by the corresponding weights $w_1,\dots ,w_5$.

Fig. 5

Training and validation accuracy of original MobileNetv3_small and modified MobileNetv3_small.

Fig. 6

Gradient values of the first eight convolutional layers in the original and modified MobileNet.

Fig. 7

Class activation maps (CAM) of CXR images in each category. Images in the first row are the original CXR images of each class, the second row lists the corresponding CAM in which the highlighted parts represent the discriminative image regions that the classifier uses to identify that category.

Fig. 8

Example CT Images of three classes [12].

Fig. 9

Validation accuracy for commonly used CNN models on the CT image dataset.

Fig. 10

The structure of the Modified ResNet18. The contents of the original ResNet18 are kept and denoted by light blue, whereas the dark blue blocks represent the modified parts. $w_1,\dots ,w_5$ are the weights respectively multiplied by the outputs of each pointwise conv block. $\times$ represents element-wise multiplication, $\oplus$ denotes the addition of five outputs of pointwise conv blocks multiplied by the corresponding weights $w_1,\dots ,w_5$.