Deep learning for COVID-19 chest CT (computed tomography) image analysis: A lesson from lung cancer

Hao Jiang; Shiming Tang; Weihuang Liu; Yang Zhang

doi:10.1016/j.csbj.2021.02.016

Deep learning for COVID-19 chest CT (computed tomography) image analysis: A lesson from lung cancer

Hao Jiang, Shiming Tang, Weihuang Liu, Yang Zhang

Research output: Contribution to journal › Article › peer-review

27 Scopus citations

Abstract

As a recent global health emergency, the quick and reliable diagnosis of COVID-19 is urgently needed. Thus, many artificial intelligence (AI)-base methods are proposed for COVID-19 chest CT (computed tomography) image analysis. However, there are very limited COVID-19 chest CT images publicly available to evaluate those deep neural networks. On the other hand, a huge amount of CT images from lung cancer are publicly available. To build a reliable deep learning model trained and tested with a larger scale dataset, the proposed model builds a public COVID-19 CT dataset, containing 1186 CT images synthesized from lung cancer CT images using CycleGAN. Additionally, various deep learning models are tested with synthesized or real chest CT images for COVID-19 and Non-COVID-19 classification. In comparison, all models achieve excellent results in accuracy, precision, recall and F1 score for both synthesized and real COVID-19 CT images, demonstrating the reliable of the synthesized dataset. The public dataset and deep learning models can facilitate the development of accurate and efficient diagnostic testing for COVID-19.

Original language	English
Pages (from-to)	1391-1399
Number of pages	9
Journal	Computational and Structural Biotechnology Journal
Volume	19
DOIs	https://doi.org/10.1016/j.csbj.2021.02.016
State	Published - Jan 2021
Externally published	Yes

Funding

The work was supported by the Shenzhen Science and Technology Innovation Commission (Shenzhen Basic Research Project No. JCYJ20180306172131515). The funders of the study had no role in data collection, analysis, interpretation, or writing of the paper. The authors had not been paid to write this article by a pharmaceutical company or other agency. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Funders	Funder number
Science, Technology and Innovation Commission of Shenzhen Municipality	JCYJ20180306172131515

Keywords

COVID-19
Chest CT image
Classification
CycleGAN
Image synthesis
Lung cancer
Style transfer

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.csbj.2021.02.016

Cite this

@article{510b70b06b8f43b5a327ce1a3b254353,

title = "Deep learning for COVID-19 chest CT (computed tomography) image analysis: A lesson from lung cancer",

abstract = "As a recent global health emergency, the quick and reliable diagnosis of COVID-19 is urgently needed. Thus, many artificial intelligence (AI)-base methods are proposed for COVID-19 chest CT (computed tomography) image analysis. However, there are very limited COVID-19 chest CT images publicly available to evaluate those deep neural networks. On the other hand, a huge amount of CT images from lung cancer are publicly available. To build a reliable deep learning model trained and tested with a larger scale dataset, the proposed model builds a public COVID-19 CT dataset, containing 1186 CT images synthesized from lung cancer CT images using CycleGAN. Additionally, various deep learning models are tested with synthesized or real chest CT images for COVID-19 and Non-COVID-19 classification. In comparison, all models achieve excellent results in accuracy, precision, recall and F1 score for both synthesized and real COVID-19 CT images, demonstrating the reliable of the synthesized dataset. The public dataset and deep learning models can facilitate the development of accurate and efficient diagnostic testing for COVID-19.",

keywords = "COVID-19, Chest CT image, Classification, CycleGAN, Image synthesis, Lung cancer, Style transfer",

author = "Hao Jiang and Shiming Tang and Weihuang Liu and Yang Zhang",

note = "Publisher Copyright: {\textcopyright} 2021 The Authors",

year = "2021",

month = jan,

doi = "10.1016/j.csbj.2021.02.016",

language = "English",

volume = "19",

pages = "1391--1399",

journal = "Computational and Structural Biotechnology Journal",

issn = "2001-0370",

publisher = "Elsevier",

}

TY - JOUR

T1 - Deep learning for COVID-19 chest CT (computed tomography) image analysis

T2 - A lesson from lung cancer

AU - Jiang, Hao

AU - Tang, Shiming

AU - Liu, Weihuang

AU - Zhang, Yang

PY - 2021/1

Y1 - 2021/1

N2 - As a recent global health emergency, the quick and reliable diagnosis of COVID-19 is urgently needed. Thus, many artificial intelligence (AI)-base methods are proposed for COVID-19 chest CT (computed tomography) image analysis. However, there are very limited COVID-19 chest CT images publicly available to evaluate those deep neural networks. On the other hand, a huge amount of CT images from lung cancer are publicly available. To build a reliable deep learning model trained and tested with a larger scale dataset, the proposed model builds a public COVID-19 CT dataset, containing 1186 CT images synthesized from lung cancer CT images using CycleGAN. Additionally, various deep learning models are tested with synthesized or real chest CT images for COVID-19 and Non-COVID-19 classification. In comparison, all models achieve excellent results in accuracy, precision, recall and F1 score for both synthesized and real COVID-19 CT images, demonstrating the reliable of the synthesized dataset. The public dataset and deep learning models can facilitate the development of accurate and efficient diagnostic testing for COVID-19.

AB - As a recent global health emergency, the quick and reliable diagnosis of COVID-19 is urgently needed. Thus, many artificial intelligence (AI)-base methods are proposed for COVID-19 chest CT (computed tomography) image analysis. However, there are very limited COVID-19 chest CT images publicly available to evaluate those deep neural networks. On the other hand, a huge amount of CT images from lung cancer are publicly available. To build a reliable deep learning model trained and tested with a larger scale dataset, the proposed model builds a public COVID-19 CT dataset, containing 1186 CT images synthesized from lung cancer CT images using CycleGAN. Additionally, various deep learning models are tested with synthesized or real chest CT images for COVID-19 and Non-COVID-19 classification. In comparison, all models achieve excellent results in accuracy, precision, recall and F1 score for both synthesized and real COVID-19 CT images, demonstrating the reliable of the synthesized dataset. The public dataset and deep learning models can facilitate the development of accurate and efficient diagnostic testing for COVID-19.

KW - COVID-19

KW - Chest CT image

KW - Classification

KW - CycleGAN

KW - Image synthesis

KW - Lung cancer

KW - Style transfer

UR - http://www.scopus.com/inward/record.url?scp=85102363403&partnerID=8YFLogxK

U2 - 10.1016/j.csbj.2021.02.016

DO - 10.1016/j.csbj.2021.02.016

M3 - Article

AN - SCOPUS:85102363403

SN - 2001-0370

VL - 19

SP - 1391

EP - 1399

JO - Computational and Structural Biotechnology Journal

JF - Computational and Structural Biotechnology Journal

ER -

Deep learning for COVID-19 chest CT (computed tomography) image analysis: A lesson from lung cancer

Abstract

Funding

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this