mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography

Sheng Wang; Jiayu Huo; Xi Ouyang; Jifei Che; Zhong Xue; Dinggang Shen; Qian Wang; Jie Zhi Cheng

doi:10.1007/978-3-030-59354-4_16

mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography

Sheng Wang, Jiayu Huo, Xi Ouyang, Jifei Che, Zhong Xue, Dinggang Shen, Qian Wang, Jie Zhi Cheng

Department of Brain and Cognitive Engineering

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

1 Citation (Scopus)

Abstract

Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learning model. In this study, we explicitly address style variety issue with the proposed multi-resolution and multi-reference neural style transfer (mr²NST) network. The mr²NST can normalize the styles from different vendors to the same style baseline with very high resolution. We illustrate that the image quality of the transferred images is comparable to the quality of original images of the target domain (vendor) in terms of NIMA scores. Meanwhile, the mr²NST results are also shown to be helpful for the lesion detection in mammograms.

Original language	English
Title of host publication	Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings
Editors	Islem Rekik, Ehsan Adeli, Sang Hyun Park, Maria del C. Valdés Hernández
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	169-177
Number of pages	9
ISBN (Print)	9783030593537
DOIs	https://doi.org/10.1007/978-3-030-59354-4_16
Publication status	Published - 2020
Event	3rd International Workshop on Predictive Intelligence in Medicine, PRIME 2020, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2020 - Lima, Peru Duration: 2020 Oct 8 → 2020 Oct 8

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12329 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	3rd International Workshop on Predictive Intelligence in Medicine, PRIME 2020, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2020
Country/Territory	Peru
City	Lima
Period	20/10/8 → 20/10/8

Bibliographical note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

Keywords

Mammography
Neural style transfer
Style normalization

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

Access to Document

10.1007/978-3-030-59354-4_16

Cite this

Wang, S., Huo, J., Ouyang, X., Che, J., Xue, Z., Shen, D., Wang, Q., & Cheng, J. Z. (2020). mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography. In I. Rekik, E. Adeli, S. H. Park, & M. D. C. Valdés Hernández (Eds.), Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings (pp. 169-177). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12329 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-59354-4_16

mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography. / Wang, Sheng; Huo, Jiayu; Ouyang, Xi et al.
Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings. ed. / Islem Rekik; Ehsan Adeli; Sang Hyun Park; Maria del C. Valdés Hernández. Springer Science and Business Media Deutschland GmbH, 2020. p. 169-177 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12329 LNCS).

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

Wang, S, Huo, J, Ouyang, X, Che, J, Xue, Z, Shen, D, Wang, Q & Cheng, JZ 2020, mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography. in I Rekik, E Adeli, SH Park & MDC Valdés Hernández (eds), Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12329 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 169-177, 3rd International Workshop on Predictive Intelligence in Medicine, PRIME 2020, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2020, Lima, Peru, 20/10/8. https://doi.org/10.1007/978-3-030-59354-4_16

Wang S, Huo J, Ouyang X, Che J, Xue Z, Shen D et al. mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography. In Rekik I, Adeli E, Park SH, Valdés Hernández MDC, editors, Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 169-177. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-59354-4_16

Wang, Sheng ; Huo, Jiayu ; Ouyang, Xi et al. / mr$$^2$$ NST : Multi-resolution and Multi-reference Neural Style Transfer for Mammography. Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings. editor / Islem Rekik ; Ehsan Adeli ; Sang Hyun Park ; Maria del C. Valdés Hernández. Springer Science and Business Media Deutschland GmbH, 2020. pp. 169-177 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{19cae4d386f7469e9631426cf422d40a,

title = "mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography",

abstract = "Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learning model. In this study, we explicitly address style variety issue with the proposed multi-resolution and multi-reference neural style transfer (mr2NST) network. The mr2NST can normalize the styles from different vendors to the same style baseline with very high resolution. We illustrate that the image quality of the transferred images is comparable to the quality of original images of the target domain (vendor) in terms of NIMA scores. Meanwhile, the mr2NST results are also shown to be helpful for the lesion detection in mammograms.",

keywords = "Mammography, Neural style transfer, Style normalization",

author = "Sheng Wang and Jiayu Huo and Xi Ouyang and Jifei Che and Zhong Xue and Dinggang Shen and Qian Wang and Cheng, {Jie Zhi}",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 3rd International Workshop on Predictive Intelligence in Medicine, PRIME 2020, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2020 ; Conference date: 08-10-2020 Through 08-10-2020",

year = "2020",

doi = "10.1007/978-3-030-59354-4_16",

language = "English",

isbn = "9783030593537",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "169--177",

editor = "Islem Rekik and Ehsan Adeli and Park, {Sang Hyun} and {Vald{\'e}s Hern{\'a}ndez}, {Maria del C.}",

booktitle = "Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings",

address = "Germany",

}

TY - GEN

T1 - mr$$^2$$ NST

T2 - 3rd International Workshop on Predictive Intelligence in Medicine, PRIME 2020, held in conjunction with the Medical Image Computing and Computer Assisted Intervention, MICCAI 2020

AU - Wang, Sheng

AU - Huo, Jiayu

AU - Ouyang, Xi

AU - Che, Jifei

AU - Xue, Zhong

AU - Shen, Dinggang

AU - Wang, Qian

AU - Cheng, Jie Zhi

PY - 2020

Y1 - 2020

N2 - Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learning model. In this study, we explicitly address style variety issue with the proposed multi-resolution and multi-reference neural style transfer (mr2NST) network. The mr2NST can normalize the styles from different vendors to the same style baseline with very high resolution. We illustrate that the image quality of the transferred images is comparable to the quality of original images of the target domain (vendor) in terms of NIMA scores. Meanwhile, the mr2NST results are also shown to be helpful for the lesion detection in mammograms.

AB - Computer-aided diagnosis with deep learning techniques has been shown to be helpful for the diagnosis of the mammography in many clinical studies. However, the image styles of different vendors are very distinctive, and there may exist domain gap among different vendors that could potentially compromise the universal applicability of one deep learning model. In this study, we explicitly address style variety issue with the proposed multi-resolution and multi-reference neural style transfer (mr2NST) network. The mr2NST can normalize the styles from different vendors to the same style baseline with very high resolution. We illustrate that the image quality of the transferred images is comparable to the quality of original images of the target domain (vendor) in terms of NIMA scores. Meanwhile, the mr2NST results are also shown to be helpful for the lesion detection in mammograms.

KW - Mammography

KW - Neural style transfer

KW - Style normalization

UR - http://www.scopus.com/inward/record.url?scp=85092921079&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-59354-4_16

DO - 10.1007/978-3-030-59354-4_16

M3 - Conference contribution

AN - SCOPUS:85092921079

SN - 9783030593537

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

SP - 169

EP - 177

BT - Predictive Intelligence in Medicine - 3rd International Workshop, PRIME 2020, Held in Conjunction with MICCAI 2020, Proceedings

A2 - Rekik, Islem

A2 - Adeli, Ehsan

A2 - Park, Sang Hyun

A2 - Valdés Hernández, Maria del C.

PB - Springer Science and Business Media Deutschland GmbH

Y2 - 8 October 2020 through 8 October 2020

ER -

mr$$^2$$ NST: Multi-resolution and Multi-reference Neural Style Transfer for Mammography

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this