SPARSE

Seed point auto-generation for random walks segmentation enhancement in medical inhomogeneous targets delineation of morphological MR and CT images

Haibin Chen, Xin Zhen, Xuejun Gu, Hao Yan, Laura Cervino, Yang Xiao, Linghong Zhou

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user-labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.

Original languageEnglish (US)
Pages (from-to)387-402
Number of pages16
JournalJournal of Applied Clinical Medical Physics
Volume16
Issue number2
DOIs
StatePublished - 2015

Fingerprint

delineation
random walk
Seed
seeds
Seeds
augmentation
Medical image processing
programming environments
weighting functions
evaluation
image processing
Labels
hardware
estimating
platforms
Hardware
Equipment and Supplies

Keywords

  • Autogeneration
  • Inhomogeneous target
  • Random walks
  • Seed point
  • Segmentation

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Radiation
  • Instrumentation

Cite this

SPARSE : Seed point auto-generation for random walks segmentation enhancement in medical inhomogeneous targets delineation of morphological MR and CT images. / Chen, Haibin; Zhen, Xin; Gu, Xuejun; Yan, Hao; Cervino, Laura; Xiao, Yang; Zhou, Linghong.

In: Journal of Applied Clinical Medical Physics, Vol. 16, No. 2, 2015, p. 387-402.

Research output: Contribution to journalArticle

@article{1032c440707d42659799a90c84bc935d,
title = "SPARSE: Seed point auto-generation for random walks segmentation enhancement in medical inhomogeneous targets delineation of morphological MR and CT images",
abstract = "In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user-labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.",
keywords = "Autogeneration, Inhomogeneous target, Random walks, Seed point, Segmentation",
author = "Haibin Chen and Xin Zhen and Xuejun Gu and Hao Yan and Laura Cervino and Yang Xiao and Linghong Zhou",
year = "2015",
doi = "10.1120/jacmp.v16i2.5324",
language = "English (US)",
volume = "16",
pages = "387--402",
journal = "Journal of Applied Clinical Medical Physics",
issn = "1526-9914",
publisher = "American Institute of Physics Publising LLC",
number = "2",

}

TY - JOUR

T1 - SPARSE

T2 - Seed point auto-generation for random walks segmentation enhancement in medical inhomogeneous targets delineation of morphological MR and CT images

AU - Chen, Haibin

AU - Zhen, Xin

AU - Gu, Xuejun

AU - Yan, Hao

AU - Cervino, Laura

AU - Xiao, Yang

AU - Zhou, Linghong

PY - 2015

Y1 - 2015

N2 - In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user-labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.

AB - In medical image processing, robust segmentation of inhomogeneous targets is a challenging problem. Because of the complexity and diversity in medical images, the commonly used semiautomatic segmentation algorithms usually fail in the segmentation of inhomogeneous objects. In this study, we propose a novel algorithm imbedded with a seed point autogeneration for random walks segmentation enhancement, namely SPARSE, for better segmentation of inhomogeneous objects. With a few user-labeled points, SPARSE is able to generate extended seed points by estimating the probability of each voxel with respect to the labels. The random walks algorithm is then applied upon the extended seed points to achieve improved segmentation result. SPARSE is implemented under the compute unified device architecture (CUDA) programming environment on graphic processing unit (GPU) hardware platform. Quantitative evaluations are performed using clinical homogeneous and inhomogeneous cases. It is found that the SPARSE can greatly decrease the sensitiveness to initial seed points in terms of location and quantity, as well as the freedom of selecting parameters in edge weighting function. The evaluation results of SPARSE also demonstrate substantial improvements in accuracy and robustness to inhomogeneous target segmentation over the original random walks algorithm.

KW - Autogeneration

KW - Inhomogeneous target

KW - Random walks

KW - Seed point

KW - Segmentation

UR - http://www.scopus.com/inward/record.url?scp=84930365696&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930365696&partnerID=8YFLogxK

U2 - 10.1120/jacmp.v16i2.5324

DO - 10.1120/jacmp.v16i2.5324

M3 - Article

VL - 16

SP - 387

EP - 402

JO - Journal of Applied Clinical Medical Physics

JF - Journal of Applied Clinical Medical Physics

SN - 1526-9914

IS - 2

ER -