Neural Denoising of Ultra-low Dose Mammography

Michael Green; Miri Sklair-Levy; Nahum Kiryati; Eli Konen; Arnaldo Mayer

doi:10.1007/978-3-030-33843-5_20

Neural Denoising of Ultra-low Dose Mammography

Michael Green, Miri Sklair-Levy, Nahum Kiryati, Eli Konen, Arnaldo Mayer

Tel Aviv University

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

Abstract

X-ray mammography is commonly used for breast cancer screening. Radiation exposure during mammography restricts the screening frequency and minimal age. Reduction of radiation dose decreases image quality. Image denoising has been recently considered as a way to facilitate dose reduction in mammography without impacting its diagnostic value. We propose a convolutional locally-consistent non-local means (CLC-NLM) algorithm for ultra-low dose mammography denoising. The proposed method achieves powerful denoising while preserving fine details in high resolution mammography. Validation is performed using a dataset of 16 digital mammography cases (4-views each). Since obtaining true low-dose and high-dose mammogram pairs raises regulatory concerns, we applied the X-ray specific and validated method of Veldkamp et al. to simulate 90% dose reduction. The proposed algorithm is shown to compete favorably, both quantitatively and qualitatively, against state-of-the-art neural denoising algorithms. In particular, tiny micro-calcifications are better preserved using the proposed algorithm.

Original language	American English
Title of host publication	Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings
Editors	Florian Knoll, Andreas Maier, Daniel Rueckert, Jong Chul Ye
Publisher	Springer
Pages	215-225
Number of pages	11
ISBN (Print)	9783030338428
DOIs	https://doi.org/10.1007/978-3-030-33843-5_20
State	Published - 1 Jan 2019
Event	2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 - Shenzhen, China Duration: 17 Oct 2019 → 17 Oct 2019

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11905 LNCS

Conference

Conference	2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Country/Territory	China
City	Shenzhen
Period	17/10/19 → 17/10/19

All Science Journal Classification (ASJC) codes

Theoretical Computer Science
General Computer Science

UN SDGs

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

Access to Document

10.1007/978-3-030-33843-5_20

Cite this

Green, M., Sklair-Levy, M., Kiryati, N., Konen, E., & Mayer, A. (2019). Neural Denoising of Ultra-low Dose Mammography. In F. Knoll, A. Maier, D. Rueckert, & J. C. Ye (Eds.), Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings (pp. 215-225). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11905 LNCS). Springer. https://doi.org/10.1007/978-3-030-33843-5_20

Green, Michael ; Sklair-Levy, Miri ; Kiryati, Nahum et al. / Neural Denoising of Ultra-low Dose Mammography. Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings. editor / Florian Knoll ; Andreas Maier ; Daniel Rueckert ; Jong Chul Ye. Springer, 2019. pp. 215-225 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{4a1a07872b334773a6968c05aaff58c2,

title = "Neural Denoising of Ultra-low Dose Mammography",

abstract = "X-ray mammography is commonly used for breast cancer screening. Radiation exposure during mammography restricts the screening frequency and minimal age. Reduction of radiation dose decreases image quality. Image denoising has been recently considered as a way to facilitate dose reduction in mammography without impacting its diagnostic value. We propose a convolutional locally-consistent non-local means (CLC-NLM) algorithm for ultra-low dose mammography denoising. The proposed method achieves powerful denoising while preserving fine details in high resolution mammography. Validation is performed using a dataset of 16 digital mammography cases (4-views each). Since obtaining true low-dose and high-dose mammogram pairs raises regulatory concerns, we applied the X-ray specific and validated method of Veldkamp et al. to simulate 90\% dose reduction. The proposed algorithm is shown to compete favorably, both quantitatively and qualitatively, against state-of-the-art neural denoising algorithms. In particular, tiny micro-calcifications are better preserved using the proposed algorithm.",

author = "Michael Green and Miri Sklair-Levy and Nahum Kiryati and Eli Konen and Arnaldo Mayer",

note = "Publisher Copyright: {\textcopyright} Springer Nature Switzerland AG 2019.; 2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019 ; Conference date: 17-10-2019 Through 17-10-2019",

year = "2019",

month = jan,

day = "1",

doi = "10.1007/978-3-030-33843-5\_20",

language = "American English",

isbn = "9783030338428",

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

publisher = "Springer",

pages = "215--225",

editor = "Florian Knoll and Andreas Maier and Daniel Rueckert and Ye, \{Jong Chul\}",

booktitle = "Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings",

address = "Germany",

}

Green, M, Sklair-Levy, M, Kiryati, N, Konen, E & Mayer, A 2019, Neural Denoising of Ultra-low Dose Mammography. in F Knoll, A Maier, D Rueckert & JC Ye (eds), Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11905 LNCS, Springer, pp. 215-225, 2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019, Shenzhen, China, 17/10/19. https://doi.org/10.1007/978-3-030-33843-5_20

Neural Denoising of Ultra-low Dose Mammography. / Green, Michael; Sklair-Levy, Miri; Kiryati, Nahum et al.
Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings. ed. / Florian Knoll; Andreas Maier; Daniel Rueckert; Jong Chul Ye. Springer, 2019. p. 215-225 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11905 LNCS).

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

TY - GEN

T1 - Neural Denoising of Ultra-low Dose Mammography

AU - Green, Michael

AU - Sklair-Levy, Miri

AU - Kiryati, Nahum

AU - Konen, Eli

AU - Mayer, Arnaldo

PY - 2019/1/1

Y1 - 2019/1/1

N2 - X-ray mammography is commonly used for breast cancer screening. Radiation exposure during mammography restricts the screening frequency and minimal age. Reduction of radiation dose decreases image quality. Image denoising has been recently considered as a way to facilitate dose reduction in mammography without impacting its diagnostic value. We propose a convolutional locally-consistent non-local means (CLC-NLM) algorithm for ultra-low dose mammography denoising. The proposed method achieves powerful denoising while preserving fine details in high resolution mammography. Validation is performed using a dataset of 16 digital mammography cases (4-views each). Since obtaining true low-dose and high-dose mammogram pairs raises regulatory concerns, we applied the X-ray specific and validated method of Veldkamp et al. to simulate 90% dose reduction. The proposed algorithm is shown to compete favorably, both quantitatively and qualitatively, against state-of-the-art neural denoising algorithms. In particular, tiny micro-calcifications are better preserved using the proposed algorithm.

AB - X-ray mammography is commonly used for breast cancer screening. Radiation exposure during mammography restricts the screening frequency and minimal age. Reduction of radiation dose decreases image quality. Image denoising has been recently considered as a way to facilitate dose reduction in mammography without impacting its diagnostic value. We propose a convolutional locally-consistent non-local means (CLC-NLM) algorithm for ultra-low dose mammography denoising. The proposed method achieves powerful denoising while preserving fine details in high resolution mammography. Validation is performed using a dataset of 16 digital mammography cases (4-views each). Since obtaining true low-dose and high-dose mammogram pairs raises regulatory concerns, we applied the X-ray specific and validated method of Veldkamp et al. to simulate 90% dose reduction. The proposed algorithm is shown to compete favorably, both quantitatively and qualitatively, against state-of-the-art neural denoising algorithms. In particular, tiny micro-calcifications are better preserved using the proposed algorithm.

UR - http://www.scopus.com/inward/record.url?scp=85076234882&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-33843-5_20

DO - 10.1007/978-3-030-33843-5_20

M3 - Conference contribution

SN - 9783030338428

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

SP - 215

EP - 225

BT - Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings

A2 - Knoll, Florian

A2 - Maier, Andreas

A2 - Rueckert, Daniel

A2 - Ye, Jong Chul

PB - Springer

T2 - 2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019

Y2 - 17 October 2019 through 17 October 2019

ER -

Green M, Sklair-Levy M, Kiryati N, Konen E, Mayer A. Neural Denoising of Ultra-low Dose Mammography. In Knoll F, Maier A, Rueckert D, Ye JC, editors, Machine Learning for Medical Image Reconstruction - 2nd International Workshop, MLMIR 2019, held in Conjunction with MICCAI 2019, Proceedings. Springer. 2019. p. 215-225. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-33843-5_20

Neural Denoising of Ultra-low Dose Mammography

Abstract

Publication series

Conference

All Science Journal Classification (ASJC) codes

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this