Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images

Chengsheng Mao, Liang Yao, Yiheng Pan, Yuan Luo, Zexian Zeng

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Thoracic diseases are serious health problems that plague a significant amount of people. Chest X-ray is currently one of the most popular methods to diagnose thoracic diseases and plays an important role in the healthcare workflow. With the success of deep learning in computer vision, a growing number of deep neural network architectures were applied to chest X-ray image classification. However, most of the previous deep neural network classifiers were based on deterministic architectures which are usually noise-sensitive and are likely to overfit the training data. In this paper, to make a deep architecture more robust to noise and to reduce overfitting, we propose using deep generative classifiers to automatically diagnose thorax diseases from the chest X-ray images. Unlike the traditional deterministic classifier, a deep generative classifier learns a distribution for each input in a middle layer of the deep neural network. A sampling layer then draws a random sample from the distribution and input it to the following layer for classification. The classifier is generative because the class label is generated from samples of a related distribution. Through training the model with a certain amount of randomness, the deep generative classifiers are expected to be robust to noise and can reduce overfitting and then achieve good performances. We implemented our deep generative classifiers based on some well-known deterministic neural network architectures and tested our models on the chest X-ray14 dataset. The results demonstrated the superiority of deep generative classifiers over the corresponding deep deterministic classifiers.

Original languageEnglish (US)
Title of host publicationProceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
EditorsHarald Schmidt, David Griol, Haiying Wang, Jan Baumbach, Huiru Zheng, Zoraida Callejas, Xiaohua Hu, Julie Dickerson, Le Zhang
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages1209-1214
Number of pages6
ISBN (Electronic)9781538654880
DOIs
StatePublished - Jan 21 2019
Event2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 - Madrid, Spain
Duration: Dec 3 2018Dec 6 2018

Publication series

NameProceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018

Conference

Conference2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018
CountrySpain
CityMadrid
Period12/3/1812/6/18

Fingerprint

Thoracic Diseases
Classifiers
Thorax
X-Rays
X rays
Noise
Plague
Workflow
Network architecture
Learning
Delivery of Health Care
Image classification
Medical problems
Health
Computer vision
Labels
Sampling
Neural networks

Keywords

  • chest-ray
  • classification
  • computer-aided diagnosis
  • deep learning
  • generative model

ASJC Scopus subject areas

  • Biomedical Engineering
  • Health Informatics

Cite this

Mao, C., Yao, L., Pan, Y., Luo, Y., & Zeng, Z. (2019). Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images. In H. Schmidt, D. Griol, H. Wang, J. Baumbach, H. Zheng, Z. Callejas, X. Hu, J. Dickerson, ... L. Zhang (Eds.), Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018 (pp. 1209-1214). [8621107] (Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM.2018.8621107
Mao, Chengsheng ; Yao, Liang ; Pan, Yiheng ; Luo, Yuan ; Zeng, Zexian. / Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images. Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018. editor / Harald Schmidt ; David Griol ; Haiying Wang ; Jan Baumbach ; Huiru Zheng ; Zoraida Callejas ; Xiaohua Hu ; Julie Dickerson ; Le Zhang. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1209-1214 (Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018).
@inproceedings{800c84a8dfa24234b3b678ee6719432b,
title = "Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images",
abstract = "Thoracic diseases are serious health problems that plague a significant amount of people. Chest X-ray is currently one of the most popular methods to diagnose thoracic diseases and plays an important role in the healthcare workflow. With the success of deep learning in computer vision, a growing number of deep neural network architectures were applied to chest X-ray image classification. However, most of the previous deep neural network classifiers were based on deterministic architectures which are usually noise-sensitive and are likely to overfit the training data. In this paper, to make a deep architecture more robust to noise and to reduce overfitting, we propose using deep generative classifiers to automatically diagnose thorax diseases from the chest X-ray images. Unlike the traditional deterministic classifier, a deep generative classifier learns a distribution for each input in a middle layer of the deep neural network. A sampling layer then draws a random sample from the distribution and input it to the following layer for classification. The classifier is generative because the class label is generated from samples of a related distribution. Through training the model with a certain amount of randomness, the deep generative classifiers are expected to be robust to noise and can reduce overfitting and then achieve good performances. We implemented our deep generative classifiers based on some well-known deterministic neural network architectures and tested our models on the chest X-ray14 dataset. The results demonstrated the superiority of deep generative classifiers over the corresponding deep deterministic classifiers.",
keywords = "chest-ray, classification, computer-aided diagnosis, deep learning, generative model",
author = "Chengsheng Mao and Liang Yao and Yiheng Pan and Yuan Luo and Zexian Zeng",
year = "2019",
month = "1",
day = "21",
doi = "10.1109/BIBM.2018.8621107",
language = "English (US)",
series = "Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "1209--1214",
editor = "Harald Schmidt and David Griol and Haiying Wang and Jan Baumbach and Huiru Zheng and Zoraida Callejas and Xiaohua Hu and Julie Dickerson and Le Zhang",
booktitle = "Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018",
address = "United States",

}

Mao, C, Yao, L, Pan, Y, Luo, Y & Zeng, Z 2019, Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images. in H Schmidt, D Griol, H Wang, J Baumbach, H Zheng, Z Callejas, X Hu, J Dickerson & L Zhang (eds), Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018., 8621107, Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018, Institute of Electrical and Electronics Engineers Inc., pp. 1209-1214, 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018, Madrid, Spain, 12/3/18. https://doi.org/10.1109/BIBM.2018.8621107

Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images. / Mao, Chengsheng; Yao, Liang; Pan, Yiheng; Luo, Yuan; Zeng, Zexian.

Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018. ed. / Harald Schmidt; David Griol; Haiying Wang; Jan Baumbach; Huiru Zheng; Zoraida Callejas; Xiaohua Hu; Julie Dickerson; Le Zhang. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1209-1214 8621107 (Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images

AU - Mao, Chengsheng

AU - Yao, Liang

AU - Pan, Yiheng

AU - Luo, Yuan

AU - Zeng, Zexian

PY - 2019/1/21

Y1 - 2019/1/21

N2 - Thoracic diseases are serious health problems that plague a significant amount of people. Chest X-ray is currently one of the most popular methods to diagnose thoracic diseases and plays an important role in the healthcare workflow. With the success of deep learning in computer vision, a growing number of deep neural network architectures were applied to chest X-ray image classification. However, most of the previous deep neural network classifiers were based on deterministic architectures which are usually noise-sensitive and are likely to overfit the training data. In this paper, to make a deep architecture more robust to noise and to reduce overfitting, we propose using deep generative classifiers to automatically diagnose thorax diseases from the chest X-ray images. Unlike the traditional deterministic classifier, a deep generative classifier learns a distribution for each input in a middle layer of the deep neural network. A sampling layer then draws a random sample from the distribution and input it to the following layer for classification. The classifier is generative because the class label is generated from samples of a related distribution. Through training the model with a certain amount of randomness, the deep generative classifiers are expected to be robust to noise and can reduce overfitting and then achieve good performances. We implemented our deep generative classifiers based on some well-known deterministic neural network architectures and tested our models on the chest X-ray14 dataset. The results demonstrated the superiority of deep generative classifiers over the corresponding deep deterministic classifiers.

AB - Thoracic diseases are serious health problems that plague a significant amount of people. Chest X-ray is currently one of the most popular methods to diagnose thoracic diseases and plays an important role in the healthcare workflow. With the success of deep learning in computer vision, a growing number of deep neural network architectures were applied to chest X-ray image classification. However, most of the previous deep neural network classifiers were based on deterministic architectures which are usually noise-sensitive and are likely to overfit the training data. In this paper, to make a deep architecture more robust to noise and to reduce overfitting, we propose using deep generative classifiers to automatically diagnose thorax diseases from the chest X-ray images. Unlike the traditional deterministic classifier, a deep generative classifier learns a distribution for each input in a middle layer of the deep neural network. A sampling layer then draws a random sample from the distribution and input it to the following layer for classification. The classifier is generative because the class label is generated from samples of a related distribution. Through training the model with a certain amount of randomness, the deep generative classifiers are expected to be robust to noise and can reduce overfitting and then achieve good performances. We implemented our deep generative classifiers based on some well-known deterministic neural network architectures and tested our models on the chest X-ray14 dataset. The results demonstrated the superiority of deep generative classifiers over the corresponding deep deterministic classifiers.

KW - chest-ray

KW - classification

KW - computer-aided diagnosis

KW - deep learning

KW - generative model

UR - http://www.scopus.com/inward/record.url?scp=85062564278&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062564278&partnerID=8YFLogxK

U2 - 10.1109/BIBM.2018.8621107

DO - 10.1109/BIBM.2018.8621107

M3 - Conference contribution

T3 - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018

SP - 1209

EP - 1214

BT - Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018

A2 - Schmidt, Harald

A2 - Griol, David

A2 - Wang, Haiying

A2 - Baumbach, Jan

A2 - Zheng, Huiru

A2 - Callejas, Zoraida

A2 - Hu, Xiaohua

A2 - Dickerson, Julie

A2 - Zhang, Le

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Mao C, Yao L, Pan Y, Luo Y, Zeng Z. Deep Generative Classifiers for Thoracic Disease Diagnosis with Chest X-ray Images. In Schmidt H, Griol D, Wang H, Baumbach J, Zheng H, Callejas Z, Hu X, Dickerson J, Zhang L, editors, Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1209-1214. 8621107. (Proceedings - 2018 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2018). https://doi.org/10.1109/BIBM.2018.8621107