DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images

Md. Rezaul Karim; Till Döhmen; Michael Cochez; Oya Beyan; Dietrich Rebholz-Schuhmann; Stefan Decker

doi:10.1109/BIBM49941.2020.9313304

DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images

Md. Rezaul Karim
, Till Döhmen
, Michael Cochez
, Oya Beyan
, Dietrich Rebholz-Schuhmann
, Stefan Decker

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

78 Scopus citations

Abstract

In this paper1, we proposed an explainable deep neural networks (DNN)-based method for automatic detection of COVID-19 symptoms from chest radiography (CXR) images, which we call 'DeepCOVIDExplainer'. We used 15,959 CXR images of 15,854 patients, covering normal, pneumonia, and COVID-19 cases. CXR images are first comprehensively preprocessed and augmented before classifying with a neural ensemble method, followed by highlighting class-discriminating regions using gradient-guided class activation maps (Grad-CAM ++) and layer-wise relevance propagation (LRP). Further, we provide human-interpretable explanations for the diagnosis. Evaluation results show that our approach can identify COVID-19 cases with a positive predictive value (PPV) of 91.6%, 92.45%, and 96.12%, respectively for normal, pneumonia, and COVID-19 cases, respectively, outperforming recent approaches.1Read longer version of this paper: https://arxiv.org/pdf/2004.04582.pdf

Original language	English
Title of host publication	2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)
Editors	Taesung Park, Young-Rae Cho, Xiaohua Tony Hu, Illhoi Yoo, Hyun Goo Woo, Jianxin Wang, Julio Facelli, Seungyoon Nam, Mingon Kang
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1034-1037
Number of pages	4
ISBN (Electronic)	9781728162157
DOIs	https://doi.org/10.1109/BIBM49941.2020.9313304
State	Published - 2020
Externally published	Yes
Event	2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 - Virtual, Seoul, Korea, Republic of Duration: Dec 16 2020 → Dec 19 2020

Publication series

Name	Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020

Conference

Conference	2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020
Country/Territory	Korea, Republic of
City	Virtual, Seoul
Period	12/16/20 → 12/19/20

Keywords

Biomedical imaging
COVID-19
Deep learning
Explainability
Grad-CAM
Layer-wise relevance propagation

Access to Document

10.1109/BIBM49941.2020.9313304

https://arxiv.org/abs/2004.04582

Cite this

Karim, M. R., Döhmen, T., Cochez, M., Beyan, O., Rebholz-Schuhmann, D., & Decker, S. (2020). DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images. In T. Park, Y.-R. Cho, X. T. Hu, I. Yoo, H. G. Woo, J. Wang, J. Facelli, S. Nam, & M. Kang (Eds.), 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (pp. 1034-1037). Article 9313304 (Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM49941.2020.9313304

Karim, Md. Rezaul ; Döhmen, Till ; Cochez, Michael et al. / DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images : Explainable COVID-19 Diagnosis from Chest X-ray Images. 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). editor / Taesung Park ; Young-Rae Cho ; Xiaohua Tony Hu ; Illhoi Yoo ; Hyun Goo Woo ; Jianxin Wang ; Julio Facelli ; Seungyoon Nam ; Mingon Kang. Institute of Electrical and Electronics Engineers Inc., 2020. pp. 1034-1037 (Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020).

@inproceedings{10221b33872944c3a03eef2daba18fb8,

title = "DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images",

abstract = "In this paper1, we proposed an explainable deep neural networks (DNN)-based method for automatic detection of COVID-19 symptoms from chest radiography (CXR) images, which we call 'DeepCOVIDExplainer'. We used 15,959 CXR images of 15,854 patients, covering normal, pneumonia, and COVID-19 cases. CXR images are first comprehensively preprocessed and augmented before classifying with a neural ensemble method, followed by highlighting class-discriminating regions using gradient-guided class activation maps (Grad-CAM ++) and layer-wise relevance propagation (LRP). Further, we provide human-interpretable explanations for the diagnosis. Evaluation results show that our approach can identify COVID-19 cases with a positive predictive value (PPV) of 91.6\%, 92.45\%, and 96.12\%, respectively for normal, pneumonia, and COVID-19 cases, respectively, outperforming recent approaches.1Read longer version of this paper: https://arxiv.org/pdf/2004.04582.pdf",

keywords = "Biomedical imaging, COVID-19, Deep learning, Explainability, Grad-CAM, Layer-wise relevance propagation",

author = "Karim, \{Md. Rezaul\} and Till D{\"o}hmen and Michael Cochez and Oya Beyan and Dietrich Rebholz-Schuhmann and Stefan Decker",

note = "Publisher Copyright: {\textcopyright} 2020 IEEE.; 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020 ; Conference date: 16-12-2020 Through 19-12-2020",

year = "2020",

doi = "10.1109/BIBM49941.2020.9313304",

language = "Ingl{\'e}s",

series = "Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1034--1037",

editor = "Taesung Park and Young-Rae Cho and Hu, \{Xiaohua Tony\} and Illhoi Yoo and Woo, \{Hyun Goo\} and Jianxin Wang and Julio Facelli and Seungyoon Nam and Mingon Kang",

booktitle = "2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)",

}

Karim, MR, Döhmen, T, Cochez, M, Beyan, O, Rebholz-Schuhmann, D & Decker, S 2020, DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images. in T Park, Y-R Cho, XT Hu, I Yoo, HG Woo, J Wang, J Facelli, S Nam & M Kang (eds), 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)., 9313304, Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020, Institute of Electrical and Electronics Engineers Inc., pp. 1034-1037, 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020, Virtual, Seoul, Korea, Republic of, 12/16/20. https://doi.org/10.1109/BIBM49941.2020.9313304

DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images. / Karim, Md. Rezaul; Döhmen, Till; Cochez, Michael et al.
2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). ed. / Taesung Park; Young-Rae Cho; Xiaohua Tony Hu; Illhoi Yoo; Hyun Goo Woo; Jianxin Wang; Julio Facelli; Seungyoon Nam; Mingon Kang. Institute of Electrical and Electronics Engineers Inc., 2020. p. 1034-1037 9313304 (Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images

T2 - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020

AU - Karim, Md. Rezaul

AU - Döhmen, Till

AU - Cochez, Michael

AU - Beyan, Oya

AU - Rebholz-Schuhmann, Dietrich

AU - Decker, Stefan

PY - 2020

Y1 - 2020

N2 - In this paper1, we proposed an explainable deep neural networks (DNN)-based method for automatic detection of COVID-19 symptoms from chest radiography (CXR) images, which we call 'DeepCOVIDExplainer'. We used 15,959 CXR images of 15,854 patients, covering normal, pneumonia, and COVID-19 cases. CXR images are first comprehensively preprocessed and augmented before classifying with a neural ensemble method, followed by highlighting class-discriminating regions using gradient-guided class activation maps (Grad-CAM ++) and layer-wise relevance propagation (LRP). Further, we provide human-interpretable explanations for the diagnosis. Evaluation results show that our approach can identify COVID-19 cases with a positive predictive value (PPV) of 91.6%, 92.45%, and 96.12%, respectively for normal, pneumonia, and COVID-19 cases, respectively, outperforming recent approaches.1Read longer version of this paper: https://arxiv.org/pdf/2004.04582.pdf

AB - In this paper1, we proposed an explainable deep neural networks (DNN)-based method for automatic detection of COVID-19 symptoms from chest radiography (CXR) images, which we call 'DeepCOVIDExplainer'. We used 15,959 CXR images of 15,854 patients, covering normal, pneumonia, and COVID-19 cases. CXR images are first comprehensively preprocessed and augmented before classifying with a neural ensemble method, followed by highlighting class-discriminating regions using gradient-guided class activation maps (Grad-CAM ++) and layer-wise relevance propagation (LRP). Further, we provide human-interpretable explanations for the diagnosis. Evaluation results show that our approach can identify COVID-19 cases with a positive predictive value (PPV) of 91.6%, 92.45%, and 96.12%, respectively for normal, pneumonia, and COVID-19 cases, respectively, outperforming recent approaches.1Read longer version of this paper: https://arxiv.org/pdf/2004.04582.pdf

KW - Biomedical imaging

KW - COVID-19

KW - Deep learning

KW - Explainability

KW - Grad-CAM

KW - Layer-wise relevance propagation

UR - https://www.scopus.com/pages/publications/85100338603

U2 - 10.1109/BIBM49941.2020.9313304

DO - 10.1109/BIBM49941.2020.9313304

M3 - Contribución a la conferencia

T3 - Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020

SP - 1034

EP - 1037

BT - 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM)

A2 - Park, Taesung

A2 - Cho, Young-Rae

A2 - Hu, Xiaohua Tony

A2 - Yoo, Illhoi

A2 - Woo, Hyun Goo

A2 - Wang, Jianxin

A2 - Facelli, Julio

A2 - Nam, Seungyoon

A2 - Kang, Mingon

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 16 December 2020 through 19 December 2020

ER -

Karim MR, Döhmen T, Cochez M, Beyan O, Rebholz-Schuhmann D, Decker S. DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images. In Park T, Cho YR, Hu XT, Yoo I, Woo HG, Wang J, Facelli J, Nam S, Kang M, editors, 2020 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). Institute of Electrical and Electronics Engineers Inc. 2020. p. 1034-1037. 9313304. (Proceedings - 2020 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2020). doi: 10.1109/BIBM49941.2020.9313304

DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

DL: ICAI Discovery Lab

Cite this

DeepCOVIDExplainer: Explainable COVID-19 Diagnosis from Chest X-ray Images: Explainable COVID-19 Diagnosis from Chest X-ray Images

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Projects

DL: ICAI Discovery Lab

Cite this