Zonal segmentation in transrectal ultrasound images of the prostate through deep learning

R. J. G. van Sloun; R. R. Wildeboer; A.W. Postema; C. K. Mannaerts; M.C.W. Gayet; H. Wijkstra; M. Mischi

doi:10.1109/ULTSYM.2018.8580157

Zonal segmentation in transrectal ultrasound images of the prostate through deep learning

R. J. G. van Sloun, R. R. Wildeboer, A.W. Postema, C. K. Mannaerts, M.C.W. Gayet, H. Wijkstra, M. Mischi

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

7 Citations (Scopus)

Abstract

Segmentation of both prostatic and zonal boundaries in transrectal ultrasound images is of great value in current clinical practice and for advancing techniques in computer-assisted diagnosis and inter-modality fusion. In this work, we propose a deep-learning approach to automatically segment the prostate and its main zones. In comparison with conventional methods, this method shows increased accuracy and Dice coefficients for full-prostate delineation. Moreover, the mean deviation between the annotated and predicted contours decreased substantially. Albeit the method still requires validation for different scanners and configurations, its real-time inference rate highlights the potential of this technique to be applied in clinical practice.

Original language	English
Title of host publication	2018 IEEE International Ultrasonics Symposium (IUS)
Place of Publication	Piscataway
Publisher	Institute of Electrical and Electronics Engineers
Number of pages	4
ISBN (Electronic)	978-1-5386-3425-7
ISBN (Print)	978-1-5386-3426-4
DOIs	https://doi.org/10.1109/ULTSYM.2018.8580157
Publication status	Published - 17 Dec 2018
Event	2018 IEEE International Ultrasonics Symposium, IUS 2018 - Kobe, Japan Duration: 22 Oct 2018 → 25 Oct 2018

Conference

Conference	2018 IEEE International Ultrasonics Symposium, IUS 2018
Abbreviated title	IUS 2018
Country/Territory	Japan
City	Kobe
Period	22/10/18 → 25/10/18

Keywords

Image segmentation
Image edge detection
Ultrasonic imaging
Magnetic resonance imaging
Medical diagnostic imaging
Biopsy
Prostate Cancer
Transrectal Ultrasound
(Auto-matic) Segmentation
Deep Learning

UN SDGs

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

Access to Document

10.1109/ULTSYM.2018.8580157

Cite this

van Sloun, R. J. G., Wildeboer, R. R., Postema, A. W., Mannaerts, C. K., Gayet, M. C. W., Wijkstra, H., & Mischi, M. (2018). Zonal segmentation in transrectal ultrasound images of the prostate through deep learning. In 2018 IEEE International Ultrasonics Symposium (IUS) Article 8580157 Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ULTSYM.2018.8580157

@inproceedings{6e93f2b85f314eb8bd9603dfb0eb5424,

title = "Zonal segmentation in transrectal ultrasound images of the prostate through deep learning",

abstract = "Segmentation of both prostatic and zonal boundaries in transrectal ultrasound images is of great value in current clinical practice and for advancing techniques in computer-assisted diagnosis and inter-modality fusion. In this work, we propose a deep-learning approach to automatically segment the prostate and its main zones. In comparison with conventional methods, this method shows increased accuracy and Dice coefficients for full-prostate delineation. Moreover, the mean deviation between the annotated and predicted contours decreased substantially. Albeit the method still requires validation for different scanners and configurations, its real-time inference rate highlights the potential of this technique to be applied in clinical practice.",

keywords = "Image segmentation, Image edge detection, Ultrasonic imaging, Magnetic resonance imaging, Medical diagnostic imaging, Biopsy, Prostate Cancer, Transrectal Ultrasound, (Auto-matic) Segmentation, Deep Learning",

author = "{van Sloun}, {R. J. G.} and Wildeboer, {R. R.} and A.W. Postema and Mannaerts, {C. K.} and M.C.W. Gayet and H. Wijkstra and M. Mischi",

year = "2018",

month = dec,

day = "17",

doi = "10.1109/ULTSYM.2018.8580157",

language = "English",

isbn = "978-1-5386-3426-4",

booktitle = "2018 IEEE International Ultrasonics Symposium (IUS)",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "2018 IEEE International Ultrasonics Symposium, IUS 2018, IUS 2018 ; Conference date: 22-10-2018 Through 25-10-2018",

}

van Sloun, RJG, Wildeboer, RR, Postema, AW, Mannaerts, CK, Gayet, MCW, Wijkstra, H & Mischi, M 2018, Zonal segmentation in transrectal ultrasound images of the prostate through deep learning. in 2018 IEEE International Ultrasonics Symposium (IUS)., 8580157, Institute of Electrical and Electronics Engineers, Piscataway, 2018 IEEE International Ultrasonics Symposium, IUS 2018, Kobe, Japan, 22/10/18. https://doi.org/10.1109/ULTSYM.2018.8580157

Zonal segmentation in transrectal ultrasound images of the prostate through deep learning. / van Sloun, R. J. G.; Wildeboer, R. R.; Postema, A.W. et al.
2018 IEEE International Ultrasonics Symposium (IUS). Piscataway: Institute of Electrical and Electronics Engineers, 2018. 8580157.

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

TY - GEN

T1 - Zonal segmentation in transrectal ultrasound images of the prostate through deep learning

AU - van Sloun, R. J. G.

AU - Wildeboer, R. R.

AU - Postema, A.W.

AU - Mannaerts, C. K.

AU - Gayet, M.C.W.

AU - Wijkstra, H.

AU - Mischi, M.

PY - 2018/12/17

Y1 - 2018/12/17

N2 - Segmentation of both prostatic and zonal boundaries in transrectal ultrasound images is of great value in current clinical practice and for advancing techniques in computer-assisted diagnosis and inter-modality fusion. In this work, we propose a deep-learning approach to automatically segment the prostate and its main zones. In comparison with conventional methods, this method shows increased accuracy and Dice coefficients for full-prostate delineation. Moreover, the mean deviation between the annotated and predicted contours decreased substantially. Albeit the method still requires validation for different scanners and configurations, its real-time inference rate highlights the potential of this technique to be applied in clinical practice.

AB - Segmentation of both prostatic and zonal boundaries in transrectal ultrasound images is of great value in current clinical practice and for advancing techniques in computer-assisted diagnosis and inter-modality fusion. In this work, we propose a deep-learning approach to automatically segment the prostate and its main zones. In comparison with conventional methods, this method shows increased accuracy and Dice coefficients for full-prostate delineation. Moreover, the mean deviation between the annotated and predicted contours decreased substantially. Albeit the method still requires validation for different scanners and configurations, its real-time inference rate highlights the potential of this technique to be applied in clinical practice.

KW - Image segmentation

KW - Image edge detection

KW - Ultrasonic imaging

KW - Magnetic resonance imaging

KW - Medical diagnostic imaging

KW - Biopsy

KW - Prostate Cancer

KW - Transrectal Ultrasound

KW - (Auto-matic) Segmentation

KW - Deep Learning

UR - http://www.scopus.com/inward/record.url?scp=85060575555&partnerID=8YFLogxK

U2 - 10.1109/ULTSYM.2018.8580157

DO - 10.1109/ULTSYM.2018.8580157

M3 - Conference contribution

SN - 978-1-5386-3426-4

BT - 2018 IEEE International Ultrasonics Symposium (IUS)

PB - Institute of Electrical and Electronics Engineers

CY - Piscataway

T2 - 2018 IEEE International Ultrasonics Symposium, IUS 2018

Y2 - 22 October 2018 through 25 October 2018

ER -

Zonal segmentation in transrectal ultrasound images of the prostate through deep learning

Abstract

Conference

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this