End-to-end lung nodule detection in computed tomography

Dufan Wu, Kyungsang Kim, Bin Dong, Georges El Fakhri, Quanzheng Li

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

2 Citations (Scopus)

Abstract

Computer aided diagnostic (CAD) system is crucial for modern medical imaging. But almost all CAD systems operate on reconstructed images, which were optimized for radiologists. Computer vision can capture features that is subtle to human observers, so it is desirable to design a CAD system operating on the raw data. In this paper, we proposed a deep-neural-network-based detection system for lung nodule detection in computed tomography (CT). A primal-dual-type deep reconstruction network was applied first to convert the raw data to the image space, followed by a 3-dimensional convolutional neural network (3D-CNN) for the nodule detection. For efficient network training, the deep reconstruction network and the CNN detector was trained sequentially first, then followed by one epoch of end-to-end fine tuning. The method was evaluated on the Lung Image Database Consortium image collection (LIDC-IDRI) with simulated forward projections. With 144 multi-slice fanbeam projections, the proposed end-to-end detector could achieve comparable sensitivity with the reference detector, which was trained and applied on the fully-sampled image data. It also demonstrated superior detection performance compared to detectors trained on the reconstructed images. The proposed method is general and could be expanded to most detection tasks in medical imaging.

Original languageEnglish (US)
Title of host publicationMachine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings
EditorsMingxia Liu, Heung-Il Suk, Yinghuan Shi
PublisherSpringer-Verlag
Pages37-45
Number of pages9
ISBN (Print)9783030009182
DOIs
StatePublished - Jan 1 2018
Externally publishedYes
Event9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018 - Granada, Spain
Duration: Sep 16 2018Sep 16 2018

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume11046 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018
CountrySpain
CityGranada
Period9/16/189/16/18

Fingerprint

Nodule
Computed Tomography
Lung
Tomography
Detectors
Detector
Medical imaging
Diagnostics
Medical Imaging
Projection
Neural Networks
Computer operating systems
Image Space
Computer vision
Primal-dual
Image Database
Tuning
Slice
Computer Vision
Convert

Keywords

  • Artificial neural networks
  • Computed tomography
  • Computer aided diagnosis

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Wu, D., Kim, K., Dong, B., Fakhri, G. E., & Li, Q. (2018). End-to-end lung nodule detection in computed tomography. In M. Liu, H-I. Suk, & Y. Shi (Eds.), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings (pp. 37-45). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS). Springer-Verlag. https://doi.org/10.1007/978-3-030-00919-9_5

End-to-end lung nodule detection in computed tomography. / Wu, Dufan; Kim, Kyungsang; Dong, Bin; Fakhri, Georges El; Li, Quanzheng.

Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. ed. / Mingxia Liu; Heung-Il Suk; Yinghuan Shi. Springer-Verlag, 2018. p. 37-45 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11046 LNCS).

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

Wu, D, Kim, K, Dong, B, Fakhri, GE & Li, Q 2018, End-to-end lung nodule detection in computed tomography. in M Liu, H-I Suk & Y Shi (eds), Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11046 LNCS, Springer-Verlag, pp. 37-45, 9th International Workshop on Machine Learning in Medical Imaging, MLMI 2018 held in conjunction with the 21st International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2018, Granada, Spain, 9/16/18. https://doi.org/10.1007/978-3-030-00919-9_5
Wu D, Kim K, Dong B, Fakhri GE, Li Q. End-to-end lung nodule detection in computed tomography. In Liu M, Suk H-I, Shi Y, editors, Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. Springer-Verlag. 2018. p. 37-45. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-030-00919-9_5
Wu, Dufan ; Kim, Kyungsang ; Dong, Bin ; Fakhri, Georges El ; Li, Quanzheng. / End-to-end lung nodule detection in computed tomography. Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings. editor / Mingxia Liu ; Heung-Il Suk ; Yinghuan Shi. Springer-Verlag, 2018. pp. 37-45 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{efceccc4fc404234967e97dd8591f40a,
title = "End-to-end lung nodule detection in computed tomography",
abstract = "Computer aided diagnostic (CAD) system is crucial for modern medical imaging. But almost all CAD systems operate on reconstructed images, which were optimized for radiologists. Computer vision can capture features that is subtle to human observers, so it is desirable to design a CAD system operating on the raw data. In this paper, we proposed a deep-neural-network-based detection system for lung nodule detection in computed tomography (CT). A primal-dual-type deep reconstruction network was applied first to convert the raw data to the image space, followed by a 3-dimensional convolutional neural network (3D-CNN) for the nodule detection. For efficient network training, the deep reconstruction network and the CNN detector was trained sequentially first, then followed by one epoch of end-to-end fine tuning. The method was evaluated on the Lung Image Database Consortium image collection (LIDC-IDRI) with simulated forward projections. With 144 multi-slice fanbeam projections, the proposed end-to-end detector could achieve comparable sensitivity with the reference detector, which was trained and applied on the fully-sampled image data. It also demonstrated superior detection performance compared to detectors trained on the reconstructed images. The proposed method is general and could be expanded to most detection tasks in medical imaging.",
keywords = "Artificial neural networks, Computed tomography, Computer aided diagnosis",
author = "Dufan Wu and Kyungsang Kim and Bin Dong and Fakhri, {Georges El} and Quanzheng Li",
year = "2018",
month = "1",
day = "1",
doi = "10.1007/978-3-030-00919-9_5",
language = "English (US)",
isbn = "9783030009182",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer-Verlag",
pages = "37--45",
editor = "Mingxia Liu and Heung-Il Suk and Yinghuan Shi",
booktitle = "Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings",

}

TY - GEN

T1 - End-to-end lung nodule detection in computed tomography

AU - Wu, Dufan

AU - Kim, Kyungsang

AU - Dong, Bin

AU - Fakhri, Georges El

AU - Li, Quanzheng

PY - 2018/1/1

Y1 - 2018/1/1

N2 - Computer aided diagnostic (CAD) system is crucial for modern medical imaging. But almost all CAD systems operate on reconstructed images, which were optimized for radiologists. Computer vision can capture features that is subtle to human observers, so it is desirable to design a CAD system operating on the raw data. In this paper, we proposed a deep-neural-network-based detection system for lung nodule detection in computed tomography (CT). A primal-dual-type deep reconstruction network was applied first to convert the raw data to the image space, followed by a 3-dimensional convolutional neural network (3D-CNN) for the nodule detection. For efficient network training, the deep reconstruction network and the CNN detector was trained sequentially first, then followed by one epoch of end-to-end fine tuning. The method was evaluated on the Lung Image Database Consortium image collection (LIDC-IDRI) with simulated forward projections. With 144 multi-slice fanbeam projections, the proposed end-to-end detector could achieve comparable sensitivity with the reference detector, which was trained and applied on the fully-sampled image data. It also demonstrated superior detection performance compared to detectors trained on the reconstructed images. The proposed method is general and could be expanded to most detection tasks in medical imaging.

AB - Computer aided diagnostic (CAD) system is crucial for modern medical imaging. But almost all CAD systems operate on reconstructed images, which were optimized for radiologists. Computer vision can capture features that is subtle to human observers, so it is desirable to design a CAD system operating on the raw data. In this paper, we proposed a deep-neural-network-based detection system for lung nodule detection in computed tomography (CT). A primal-dual-type deep reconstruction network was applied first to convert the raw data to the image space, followed by a 3-dimensional convolutional neural network (3D-CNN) for the nodule detection. For efficient network training, the deep reconstruction network and the CNN detector was trained sequentially first, then followed by one epoch of end-to-end fine tuning. The method was evaluated on the Lung Image Database Consortium image collection (LIDC-IDRI) with simulated forward projections. With 144 multi-slice fanbeam projections, the proposed end-to-end detector could achieve comparable sensitivity with the reference detector, which was trained and applied on the fully-sampled image data. It also demonstrated superior detection performance compared to detectors trained on the reconstructed images. The proposed method is general and could be expanded to most detection tasks in medical imaging.

KW - Artificial neural networks

KW - Computed tomography

KW - Computer aided diagnosis

UR - http://www.scopus.com/inward/record.url?scp=85054499052&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85054499052&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-00919-9_5

DO - 10.1007/978-3-030-00919-9_5

M3 - Conference contribution

AN - SCOPUS:85054499052

SN - 9783030009182

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 37

EP - 45

BT - Machine Learning in Medical Imaging - 9th International Workshop, MLMI 2018, Held in Conjunction with MICCAI 2018, Proceedings

A2 - Liu, Mingxia

A2 - Suk, Heung-Il

A2 - Shi, Yinghuan

PB - Springer-Verlag

ER -