Convolutional neural networks for prostate magnetic resonance image segmentation

Tahereh Hassanzadeh; Leonard G. C. Hamey; Kevin Ho-Shon

doi:10.1109/ACCESS.2019.2903284

Convolutional neural networks for prostate magnetic resonance image segmentation

Tahereh Hassanzadeh^*, Leonard G. C. Hamey, Kevin Ho-Shon

^*Corresponding author for this work

School of Computing

Research output: Contribution to journal › Article › peer-review

32 Citations (Scopus)

66 Downloads (Pure)

Abstract

One of the most accurate and non-invasive prostate imaging methods is magnetic resonance imaging (MRI). Segmentation is needed to find the boundary of the prostate, either automatically or semi-automatically. Recently, fully convolutional neural networks (FCNN) are being used for this purpose. In this paper, to improve the FCNN performance for prostate MRI segmentation, we analyze various structures of shortcut connections together with the size of a deep network and suggest eight different FCNNs-based deep 2D network structures for automatic MRI prostate segmentation. Our evaluations on the PROMISE12 dataset with ten-fold cross-validation indicate improved and competitive results. We analyze the results in detail, considering MRI slices, MRI volumes, test folds, and also the impact on prostate segmentation of using an EndoRectal Coil to capture the prostate MRI. Our best 2D network outperforms the state-of-the-art 3D FCNN-based methods for prostate MRI segmentation on publicly available data, without any further post-processing.

Original language	English
Article number	8666973
Pages (from-to)	36748-36760
Number of pages	13
Journal	IEEE Access
Volume	7
DOIs	https://doi.org/10.1109/ACCESS.2019.2903284
Publication status	Published - 13 Mar 2019

Bibliographical note

Copyright the Author(s) 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Keywords

Image segmentation
Magnetic resonance imaging
Feature extraction
Prostate cancer
Convolutional Neural Network
Automatic MRI segmentation
Fully Convolutional Neural Network
Prostate MRI segmentation

Access to Document

10.1109/ACCESS.2019.2903284

Publisher version (open access)Final published version, 7.39 MBLicence: CC BY

Cite this

@article{0bbdd4b1b6ba4670a172da82fc9135f1,

title = "Convolutional neural networks for prostate magnetic resonance image segmentation",

abstract = "One of the most accurate and non-invasive prostate imaging methods is magnetic resonance imaging (MRI). Segmentation is needed to find the boundary of the prostate, either automatically or semi-automatically. Recently, fully convolutional neural networks (FCNN) are being used for this purpose. In this paper, to improve the FCNN performance for prostate MRI segmentation, we analyze various structures of shortcut connections together with the size of a deep network and suggest eight different FCNNs-based deep 2D network structures for automatic MRI prostate segmentation. Our evaluations on the PROMISE12 dataset with ten-fold cross-validation indicate improved and competitive results. We analyze the results in detail, considering MRI slices, MRI volumes, test folds, and also the impact on prostate segmentation of using an EndoRectal Coil to capture the prostate MRI. Our best 2D network outperforms the state-of-the-art 3D FCNN-based methods for prostate MRI segmentation on publicly available data, without any further post-processing.",

keywords = "Image segmentation, Magnetic resonance imaging, Feature extraction, Prostate cancer, Convolutional Neural Network, Automatic MRI segmentation, Fully Convolutional Neural Network, Prostate MRI segmentation",

author = "Tahereh Hassanzadeh and Hamey, {Leonard G. C.} and Kevin Ho-Shon",

note = "Copyright the Author(s) 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher. ",

year = "2019",

month = mar,

day = "13",

doi = "10.1109/ACCESS.2019.2903284",

language = "English",

volume = "7",

pages = "36748--36760",

journal = "IEEE Access",

issn = "2169-3536",

publisher = "Institute of Electrical and Electronics Engineers (IEEE)",

}

TY - JOUR

T1 - Convolutional neural networks for prostate magnetic resonance image segmentation

AU - Hassanzadeh, Tahereh

AU - Hamey, Leonard G. C.

AU - Ho-Shon, Kevin

N1 - Copyright the Author(s) 2019. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2019/3/13

Y1 - 2019/3/13

N2 - One of the most accurate and non-invasive prostate imaging methods is magnetic resonance imaging (MRI). Segmentation is needed to find the boundary of the prostate, either automatically or semi-automatically. Recently, fully convolutional neural networks (FCNN) are being used for this purpose. In this paper, to improve the FCNN performance for prostate MRI segmentation, we analyze various structures of shortcut connections together with the size of a deep network and suggest eight different FCNNs-based deep 2D network structures for automatic MRI prostate segmentation. Our evaluations on the PROMISE12 dataset with ten-fold cross-validation indicate improved and competitive results. We analyze the results in detail, considering MRI slices, MRI volumes, test folds, and also the impact on prostate segmentation of using an EndoRectal Coil to capture the prostate MRI. Our best 2D network outperforms the state-of-the-art 3D FCNN-based methods for prostate MRI segmentation on publicly available data, without any further post-processing.

AB - One of the most accurate and non-invasive prostate imaging methods is magnetic resonance imaging (MRI). Segmentation is needed to find the boundary of the prostate, either automatically or semi-automatically. Recently, fully convolutional neural networks (FCNN) are being used for this purpose. In this paper, to improve the FCNN performance for prostate MRI segmentation, we analyze various structures of shortcut connections together with the size of a deep network and suggest eight different FCNNs-based deep 2D network structures for automatic MRI prostate segmentation. Our evaluations on the PROMISE12 dataset with ten-fold cross-validation indicate improved and competitive results. We analyze the results in detail, considering MRI slices, MRI volumes, test folds, and also the impact on prostate segmentation of using an EndoRectal Coil to capture the prostate MRI. Our best 2D network outperforms the state-of-the-art 3D FCNN-based methods for prostate MRI segmentation on publicly available data, without any further post-processing.

KW - Image segmentation

KW - Magnetic resonance imaging

KW - Feature extraction

KW - Prostate cancer

KW - Convolutional Neural Network

KW - Automatic MRI segmentation

KW - Fully Convolutional Neural Network

KW - Prostate MRI segmentation

UR - http://www.scopus.com/inward/record.url?scp=85065166485&partnerID=8YFLogxK

U2 - 10.1109/ACCESS.2019.2903284

DO - 10.1109/ACCESS.2019.2903284

M3 - Article

AN - SCOPUS:85065166485

SN - 2169-3536

VL - 7

SP - 36748

EP - 36760

JO - IEEE Access

JF - IEEE Access

M1 - 8666973

ER -

Convolutional neural networks for prostate magnetic resonance image segmentation

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this