Convolutional Neural Networks with Transfer Learning for Recognition of COVID-19: A Comparative Study of Different Approaches

Tanmay Garg,Mamta Garg,Akhil Ranjan Garg,Om Prakash Mahela

doi:10.3390/ai1040034

Abstract

To judge the ability of convolutional neural networks (CNNs) to effectively and efficiently transfer image representations learned on the ImageNet dataset to the task of recognizing COVID-19 in this work, we propose and analyze four approaches. For this purpose, we use VGG16, ResNetV2, InceptionResNetV2, DenseNet121, and MobileNetV2 CNN models pre-trained on ImageNet dataset to extract features from X-ray images of COVID and Non-COVID patients. Simulations study performed by us reveal that these pre-trained models have a different level of ability to transfer image representation. We find that in the approaches that we have proposed, if we use either ResNetV2 or DenseNet121 to extract features, then the performance of these approaches to detect COVID-19 is better. One of the important findings of our study is that the use of principal component analysis for feature selection improves efficiency. The approach using the fusion of features outperforms all the other approaches, and with this approach, we could achieve an accuracy of 0.94 for a three-class classification problem. This work will not only be useful for COVID-19 detection but also for any domain with small datasets.

Highlights

COVID-19, a global pandemic, is still spreading in many parts of the world since its identification in late December 2019
Several models have performed exceedingly well on the ImageNet data classification task [25]. We evaluate five such deep networks, namely VGG16, ResNet50V2, InceptionResNetV2, DenseNet121, and MobileNetV2, and compare their performance when used in transfer learning mode for detection of COVID-19
Simulation results for different approaches discussed in the previous sections use pre-trained convolutional neural networks (CNNs)

Summary

Introduction

COVID-19, a global pandemic, is still spreading in many parts of the world since its identification in late December 2019. In these nine to ten months, this disease has become one of the most significant public health emergencies requiring remedial measures and early diagnosis. In many countries till recently, reverse transcription-polymerase chain reaction (RT-PCR) tests are the most popular diagnostic method for detecting COVID-19. Popular, this method suffers from limitations in its long wait time and low sensitivity. Both the RT-PCR method or molecular testing approach need expensive equipment and trained professionals

Objectives

Methods

Results

Conclusion