Superpixel-Based Segmentation for 3D Prostate MR Images

Zhiqiang Tian, Lizhi Liu, Zhenfeng Zhang, Baowei Fei

Research output: Contribution to journalArticle

50 Citations (Scopus)

Abstract

This paper proposes a method for segmenting the prostate on magnetic resonance (MR) images. A superpixel-based 3D graph cut algorithm is proposed to obtain the prostate surface. Instead of pixels, superpixels are considered as the basic processing units to construct a 3D superpixel-based graph. The superpixels are labeled as the prostate or background by minimizing an energy function using graph cut based on the 3D superpixel-based graph. To construct the energy function, we proposed a superpixel-based shape data term, an appearance data term, and two superpixel-based smoothness terms. The proposed superpixel-based terms provide the effectiveness and robustness for the segmentation of the prostate. The segmentation result of graph cuts is used as an initialization of a 3D active contour model to overcome the drawback of the graph cut. The result of 3D active contour model is then used to update the shape model and appearance model of the graph cut. Iterations of the 3D graph cut and 3D active contour model have the ability to jump out of local minima and obtain a smooth prostate surface. On our 43 MR volumes, the proposed method yields a mean Dice ratio of 89.3 ± 1.9%. On PROMISE12 test data set, our method was ranked at the second place; the mean Dice ratio and standard deviation is 87.0 ± 3.2%. The experimental results show that the proposed method outperforms several state-of-the-art prostate MRI segmentation methods.

Original languageEnglish (US)
Article number7312972
Pages (from-to)791-801
Number of pages11
JournalIEEE Transactions on Medical Imaging
Volume35
Issue number3
DOIs
StatePublished - Mar 2016
Externally publishedYes

Fingerprint

Magnetic resonance
Prostate
Magnetic Resonance Spectroscopy
Magnetic resonance imaging
Pixels
Processing

Keywords

  • 3D graph cuts
  • active contour model
  • magnetic resonance imaging (MRI)
  • Prostate segmentation
  • superpixel

ASJC Scopus subject areas

  • Software
  • Radiological and Ultrasound Technology
  • Computer Science Applications
  • Electrical and Electronic Engineering

Cite this

Superpixel-Based Segmentation for 3D Prostate MR Images. / Tian, Zhiqiang; Liu, Lizhi; Zhang, Zhenfeng; Fei, Baowei.

In: IEEE Transactions on Medical Imaging, Vol. 35, No. 3, 7312972, 03.2016, p. 791-801.

Research output: Contribution to journalArticle

Tian, Zhiqiang ; Liu, Lizhi ; Zhang, Zhenfeng ; Fei, Baowei. / Superpixel-Based Segmentation for 3D Prostate MR Images. In: IEEE Transactions on Medical Imaging. 2016 ; Vol. 35, No. 3. pp. 791-801.
@article{42c3ba0424df4a67b994d93563b79b95,
title = "Superpixel-Based Segmentation for 3D Prostate MR Images",
abstract = "This paper proposes a method for segmenting the prostate on magnetic resonance (MR) images. A superpixel-based 3D graph cut algorithm is proposed to obtain the prostate surface. Instead of pixels, superpixels are considered as the basic processing units to construct a 3D superpixel-based graph. The superpixels are labeled as the prostate or background by minimizing an energy function using graph cut based on the 3D superpixel-based graph. To construct the energy function, we proposed a superpixel-based shape data term, an appearance data term, and two superpixel-based smoothness terms. The proposed superpixel-based terms provide the effectiveness and robustness for the segmentation of the prostate. The segmentation result of graph cuts is used as an initialization of a 3D active contour model to overcome the drawback of the graph cut. The result of 3D active contour model is then used to update the shape model and appearance model of the graph cut. Iterations of the 3D graph cut and 3D active contour model have the ability to jump out of local minima and obtain a smooth prostate surface. On our 43 MR volumes, the proposed method yields a mean Dice ratio of 89.3 ± 1.9{\%}. On PROMISE12 test data set, our method was ranked at the second place; the mean Dice ratio and standard deviation is 87.0 ± 3.2{\%}. The experimental results show that the proposed method outperforms several state-of-the-art prostate MRI segmentation methods.",
keywords = "3D graph cuts, active contour model, magnetic resonance imaging (MRI), Prostate segmentation, superpixel",
author = "Zhiqiang Tian and Lizhi Liu and Zhenfeng Zhang and Baowei Fei",
year = "2016",
month = "3",
doi = "10.1109/TMI.2015.2496296",
language = "English (US)",
volume = "35",
pages = "791--801",
journal = "IEEE Transactions on Medical Imaging",
issn = "0278-0062",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "3",

}

TY - JOUR

T1 - Superpixel-Based Segmentation for 3D Prostate MR Images

AU - Tian, Zhiqiang

AU - Liu, Lizhi

AU - Zhang, Zhenfeng

AU - Fei, Baowei

PY - 2016/3

Y1 - 2016/3

N2 - This paper proposes a method for segmenting the prostate on magnetic resonance (MR) images. A superpixel-based 3D graph cut algorithm is proposed to obtain the prostate surface. Instead of pixels, superpixels are considered as the basic processing units to construct a 3D superpixel-based graph. The superpixels are labeled as the prostate or background by minimizing an energy function using graph cut based on the 3D superpixel-based graph. To construct the energy function, we proposed a superpixel-based shape data term, an appearance data term, and two superpixel-based smoothness terms. The proposed superpixel-based terms provide the effectiveness and robustness for the segmentation of the prostate. The segmentation result of graph cuts is used as an initialization of a 3D active contour model to overcome the drawback of the graph cut. The result of 3D active contour model is then used to update the shape model and appearance model of the graph cut. Iterations of the 3D graph cut and 3D active contour model have the ability to jump out of local minima and obtain a smooth prostate surface. On our 43 MR volumes, the proposed method yields a mean Dice ratio of 89.3 ± 1.9%. On PROMISE12 test data set, our method was ranked at the second place; the mean Dice ratio and standard deviation is 87.0 ± 3.2%. The experimental results show that the proposed method outperforms several state-of-the-art prostate MRI segmentation methods.

AB - This paper proposes a method for segmenting the prostate on magnetic resonance (MR) images. A superpixel-based 3D graph cut algorithm is proposed to obtain the prostate surface. Instead of pixels, superpixels are considered as the basic processing units to construct a 3D superpixel-based graph. The superpixels are labeled as the prostate or background by minimizing an energy function using graph cut based on the 3D superpixel-based graph. To construct the energy function, we proposed a superpixel-based shape data term, an appearance data term, and two superpixel-based smoothness terms. The proposed superpixel-based terms provide the effectiveness and robustness for the segmentation of the prostate. The segmentation result of graph cuts is used as an initialization of a 3D active contour model to overcome the drawback of the graph cut. The result of 3D active contour model is then used to update the shape model and appearance model of the graph cut. Iterations of the 3D graph cut and 3D active contour model have the ability to jump out of local minima and obtain a smooth prostate surface. On our 43 MR volumes, the proposed method yields a mean Dice ratio of 89.3 ± 1.9%. On PROMISE12 test data set, our method was ranked at the second place; the mean Dice ratio and standard deviation is 87.0 ± 3.2%. The experimental results show that the proposed method outperforms several state-of-the-art prostate MRI segmentation methods.

KW - 3D graph cuts

KW - active contour model

KW - magnetic resonance imaging (MRI)

KW - Prostate segmentation

KW - superpixel

UR - http://www.scopus.com/inward/record.url?scp=84963748030&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84963748030&partnerID=8YFLogxK

U2 - 10.1109/TMI.2015.2496296

DO - 10.1109/TMI.2015.2496296

M3 - Article

C2 - 26540678

AN - SCOPUS:84963748030

VL - 35

SP - 791

EP - 801

JO - IEEE Transactions on Medical Imaging

JF - IEEE Transactions on Medical Imaging

SN - 0278-0062

IS - 3

M1 - 7312972

ER -