A Survey on Machine Learning in COVID-19 Diagnosis

Yu-Dong Zhang,Xing Guo,Siyuan Lu,Zhihai Lu

doi:10.32604/cmes.2022.017679

Abstract

Since Corona Virus Disease 2019 outbreak, many expert groups worldwide have studied the problem and proposed many diagnostic methods. This paper focuses on the research of Corona Virus Disease 2019 diagnosis. First, the procedure of the diagnosis based on machine learning is introduced in detail, which includes medical data collection, image preprocessing, feature extraction, and image classification. Then, we review seven methods in detail: transfer learning, ensemble learning, unsupervised learning and semi-supervised learning, convolutional neural networks, graph neural networks, explainable deep neural networks, and so on. What’s more, the advantages and limitations of different diagnosis methods are compared. Although the great achievements in medical images classification in recent years, Corona Virus Disease 2019 images classification based on machine learning still encountered many problems. For example, the highly unbalanced dataset, the difficulty of collecting labeled data, and the poor quality of the data. Aiming at these problems, we propose some solutions and provide a comprehensive presentation for future research.

Highlights

The novel coronavirus pneumonia broke out in 2019 [1]
The novel coronavirus pneumonia is transmitted through interpersonal transmission [3], and the recent emergence of large numbers of infected people without initial symptoms of infection accelerates the spread of the disease [4], the surge in patients has put a lot of pressure on medical institutions [5]
3.2 COVID-19 Diagnosis Based on Convolutional Neural Networks Training from Scratch convolutional neural networks (CNNs) play an important role in clinical diagnosis

Summary

Introduction

The novel coronavirus pneumonia broke out in 2019 [1]. The pathogen is identified as a new enveloped ribonucleic acid-β (RNA-β) coronavirus, and it is similar to Severe Acute Respiratory Syndrome Coronavirus (SARS-CoV), named SARS-CoV-2 [2]. Through reading the literature about the diagnosis of the COVID-19, it was found that the following seven learning methods (transfer learning, ensemble learning, unsupervised learning and semi-supervised learning, convolutional neural networks, graph neural networks, explainable deep neural networks, and so on) are more commonly used in the diagnosis of the COVID-19. Most experiments encountered some problems, such as the highly unbalanced dataset, the difficulty of collecting labeled data, and the poor quality of the data We summarize these seven learning methods in this paper, and propose solutions to these problems encountered. The method can be used as an aid in the decision-making process of clinicians and reduce doctors’ stress in disease diagnosis [22], when COVID-19 breaks out in the world in a short time. Extreme learning machine Area under curve Class activation mapping Fully connected

Medical Data

Images Preprocessing

Results

Features Extraction

Classification

COVID-19 Diagnosis Based on Transfer Learning

COVID-19 Diagnosis Based on Convolutional Neural Networks Training from

COVID-19 Diagnosis Based on Ensemble Learning

COVID-19 Diagnosis Based on Unsupervised Learning and Semi-Supervised Learning

COVID-19 Diagnosis Based on Graph Neural Networks

COVID-19 Diagnosis Based on Explainable Deep Neural Networks

Limitations and Conclusion