A Deep Learning approach to detect COVID-19

In this paper, we propose a rapidly testing method which has a high productivity in a short time. In details, we will apply deep learning neural networks, e.g. ResNet50 and VGG19 to solve this problem. After that, we will proceed analysing pros and cons of those models for a thorough vision about applying artificial intelligence in COVID-19 rapid testing.

Contents

1. Introduction
2. Prerequisites
3. Repo structure
4. Install and set up environment
5. Results
6. References

1. Introduction

This work is conducted by:

• Tan Ngoc Pham
• An Vo

Dataset's description

In this research, we use COVIDx dataset [2] - which is a widely used dataset in recent research about COVID-19 nowadays. COVIDx Datset is a dataset synthesized from many a different source, which are in details: Cohen et al. [3], Chung [4], Chung [5], Radiological Society of North America [5],and Radiological Society of North America [6]. Additionally, this dataset also provides an image extension transfer tool: from .mri into .jpg. And the author moreover provide a code to support data pre-processing and getting rid of unnecessary part for synthesized data.

Method

We proposed using diagnostic imaging, which is an approach using Chest X-ray (CXR) image. This is due to its lower cost and faster testing time in comparison with Real-time Polymerase Chain Reaction (RT PCR) or Computed Tomography (CT) Image.

Model approach

Throughout this research, we use 2 different approaches which are ResNet50 and VGG19 to solve this problem. VGG19 is a deep neural network architecture under-using residual design principals, it is also a compact architecture which has a low diversity of architectures. On the other hand, ResNet50 is a deep neural network harnessing residual design principles and it has a moderate diversity of architectures. This network brings many a high productivity in a large number of researching in classifying X-ray images. Despite each approach has its own benefits and drawbacks, both are proved their productivity through real application.

2. Prerequisites

• Python >= 3.6
• Sklearn >= 0.24.2
• NumPy >= 1.13.3
• Tensorflow >= 2.6

3. Repo structure

• Model: deep convolutional neural network architectures
  • VGG19
  • ResNet50 - 14 epochs
  • ResNet50 - 50 epochs
• Plot: data plot for each architecture
  • VGG19
    • Loss for training
    • Model accuracy
    • Confusion matrix
  • ResNet50 (14 epochs)
    • Loss for training
    • Model accuracy
    • Confusion matrix
  • ResNet50 (50 epochs)
    • Loss for training
    • Model accuracy
    • Confusion matrix
  • Accuracy on training set
  • Accuracy on validation set
  • Loss for training set
  • Accuracy on validation set
• Report: detailed report for this research
  • src: LaTeX source for report
  • Detect COVID-19 with Pneumonia via DCNNs.pdf: publication's name
  • Presentation.pptx: presentation file
• Demo.ipynb: Demo to prototyping on jupyter notebook

4. Install and set up environment

conda create -n covid-detect python==3.8 -y
conda activate covid-detect 
pip3 install -r requirements.txt
jupyter notebook

5. Results

Results on VGG19

Disease	Precision	Recall	F1-score	Support
COVID-19	0.99	0.82	0.90	274
Non-respiratory disease	0.7	0.96	0.81	100
Pneumonia	0.8	0.86	0.83	105

Results on ResNet50 (14 epochs)

Disease	Precision	Recall	F1-score	Support
COVID-19	0.97	0.67	0.79	274
Non-respiratory disease	0.56	0.96	0.71	100
Pneumonia	0.74	0.85	0.79	105

Results on ResNet50 (50 epochs)

Disease	Precision	Recall	F1-score	Support
COVID-19	0.96	0.80	0.88	274
Non-respiratory disease	0.73	0.86	0.79	100
Pneumonia	0.71	0.90	0.79	105

Comparison among models based on sensitivity

Architecture	Non-respiratory disease	Pneumonia	COVID-19
VGG19	96%	86%	82%
ResNet50 (14 epochs)	96%	85%	67%
ResNet50 (50 epochs)	86%	90%	80%

Comparison among models based on PPV

Architecture	Non-respiratory disease	Pneumonia	COVID-19
VGG19	70%	80%	99%
ResNet50 (14 epochs)	56%	74%	97%
ResNet50 (50 epochs)	73%	71%	96%

Comparison between precision and number of parameters among models

Architecture	Number of parameters (M)	Accuracy	Resolution
VGG19	29.76 trainable + 20.25 non-trainable	86%	480 x 480
ResNet50 (14 epochs)	25.93 trainable + 23.77 non-trainable	77%	224 x 224
ResNet50 (50 epochs)	25.93 trainable + 23.77 non-trainable	84%	224 x 224

6. References

• [1] Cleverley, J.; Piper, J.; Jones, M.M. The role of chest radiography in confirming covid-19 pneumonia. BMJ 2020, 370.
• [2] Wang, L., Lin, Z.Q. & Wong, A. COVID-Net: a tailored deep convolutional neural network design for detection of COVID-19 cases from chest X-ray images. Sci Rep 10, 19549 (2020). https://doi.org/10.1038/s41598-020-76550-z
• [3] Cohen, J. P., Morrison, P., & Dao, L. COVID-19 image data collection. arXiv:2003.11597(2020)
• [4] Chung, A. Figure 1 COVID-19 chest x-ray data initiative. https://github.com/agchung/Figure1-COVID-chestxray-dataset (2020)
• [5] Chung, A. Actualmed COVID-19 chest x-ray data initiative. https://github.com/agchung/Actualmed-COVID-chestxray-dataset (2020)
• [6] Radiological Society of North America. COVID-19 radiography database. https://www.kaggle.com/tawsifurrahman/covid19-radiography-database (2019)
• [7] Radiological Society of North America. RSNA pneumonia detection challenge. https://www.kaggle.com/c/rsna-pneumonia-detection-challenge/data (2019).
• [8] Krizhevsky, A., Sutskever, I. & Hinton, G. E. (2012). ImageNet Classification with Deep Convolutional Neural Networks. In F. Pereira, C. J. C. Burges, L. Bottou & K. Q. Weinberger (ed.), Advances in Neural Information Processing Systems 25 (pp. 1097--1105) . Curran Associates, Inc. . 
• [9] K. Simonyan et al., Very deep convolutional networks for large-scale image recognition, 2015.