Correction of inhomogeneous magnetic resonance images using multiscale retinex for segmentation accuracy improvement

Wen Hung Chao, Hsin Yi Lai, Yen Yu I Shih, You Yin Chen, Yu Chun Lo, Sheng Huang Lin, Siny Tsang, Robby Wu, Fu Shan Jaw

Research output: Contribution to journalArticle

9 Citations (Scopus)

Abstract

The purpose of this study was to improve the accuracy of tissue segmentation on brain magnetic resonance (MR) images preprocessed by multiscale retinex (MSR), segmented with a combined boosted decision tree (BDT) and MSR algorithm (hereinafter referred to as the MSRBDT algorithm). Simulated brain MR (SBMR) T1-weighted images of different noise levels and RF inhomogeneities were adopted to evaluate the outcome of the proposed method; the MSRBDT algorithm was used to identify the gray matter (GM), white matter (WM), and cerebral-spinal fluid (CSF) in the brain tissues. The accuracy rates of GM, WM, and CSF segmentation, with spatial features (G, x, y, r, θ), were respectively greater than 0.9805, 0.9817, and 0.9871. In addition, images segmented with the MSRBDT algorithm were better than those obtained with the expectation maximization (EM) algorithm; brain tissue segmentation in MR images was significantly more precise. The proposed MSRBDT algorithm could be beneficial in clinical image segmentation.

Original languageEnglish
Pages (from-to)129-140
Number of pages12
JournalBiomedical Signal Processing and Control
Volume7
Issue number2
DOIs
Publication statusPublished - Mar 2012
Externally publishedYes

Fingerprint

Magnetic resonance
Magnetic Resonance Spectroscopy
Brain
Tissue
Decision Trees
Fluids
Decision trees
Image segmentation
Noise

Keywords

  • Boosted decision tree
  • Brain tissue
  • Multiscale retinex
  • Segmentation
  • Spatial feature

ASJC Scopus subject areas

  • Health Informatics
  • Signal Processing

Cite this

Correction of inhomogeneous magnetic resonance images using multiscale retinex for segmentation accuracy improvement. / Chao, Wen Hung; Lai, Hsin Yi; Shih, Yen Yu I; Chen, You Yin; Lo, Yu Chun; Lin, Sheng Huang; Tsang, Siny; Wu, Robby; Jaw, Fu Shan.

In: Biomedical Signal Processing and Control, Vol. 7, No. 2, 03.2012, p. 129-140.

Research output: Contribution to journalArticle

Chao, Wen Hung ; Lai, Hsin Yi ; Shih, Yen Yu I ; Chen, You Yin ; Lo, Yu Chun ; Lin, Sheng Huang ; Tsang, Siny ; Wu, Robby ; Jaw, Fu Shan. / Correction of inhomogeneous magnetic resonance images using multiscale retinex for segmentation accuracy improvement. In: Biomedical Signal Processing and Control. 2012 ; Vol. 7, No. 2. pp. 129-140.
@article{74c72469bcde41d7ab812c0ed8e5f642,
title = "Correction of inhomogeneous magnetic resonance images using multiscale retinex for segmentation accuracy improvement",
abstract = "The purpose of this study was to improve the accuracy of tissue segmentation on brain magnetic resonance (MR) images preprocessed by multiscale retinex (MSR), segmented with a combined boosted decision tree (BDT) and MSR algorithm (hereinafter referred to as the MSRBDT algorithm). Simulated brain MR (SBMR) T1-weighted images of different noise levels and RF inhomogeneities were adopted to evaluate the outcome of the proposed method; the MSRBDT algorithm was used to identify the gray matter (GM), white matter (WM), and cerebral-spinal fluid (CSF) in the brain tissues. The accuracy rates of GM, WM, and CSF segmentation, with spatial features (G, x, y, r, θ), were respectively greater than 0.9805, 0.9817, and 0.9871. In addition, images segmented with the MSRBDT algorithm were better than those obtained with the expectation maximization (EM) algorithm; brain tissue segmentation in MR images was significantly more precise. The proposed MSRBDT algorithm could be beneficial in clinical image segmentation.",
keywords = "Boosted decision tree, Brain tissue, Multiscale retinex, Segmentation, Spatial feature",
author = "Chao, {Wen Hung} and Lai, {Hsin Yi} and Shih, {Yen Yu I} and Chen, {You Yin} and Lo, {Yu Chun} and Lin, {Sheng Huang} and Siny Tsang and Robby Wu and Jaw, {Fu Shan}",
year = "2012",
month = "3",
doi = "10.1016/j.bspc.2011.04.001",
language = "English",
volume = "7",
pages = "129--140",
journal = "Biomedical Signal Processing and Control",
issn = "1746-8094",
publisher = "Elsevier BV",
number = "2",

}

TY - JOUR

T1 - Correction of inhomogeneous magnetic resonance images using multiscale retinex for segmentation accuracy improvement

AU - Chao, Wen Hung

AU - Lai, Hsin Yi

AU - Shih, Yen Yu I

AU - Chen, You Yin

AU - Lo, Yu Chun

AU - Lin, Sheng Huang

AU - Tsang, Siny

AU - Wu, Robby

AU - Jaw, Fu Shan

PY - 2012/3

Y1 - 2012/3

N2 - The purpose of this study was to improve the accuracy of tissue segmentation on brain magnetic resonance (MR) images preprocessed by multiscale retinex (MSR), segmented with a combined boosted decision tree (BDT) and MSR algorithm (hereinafter referred to as the MSRBDT algorithm). Simulated brain MR (SBMR) T1-weighted images of different noise levels and RF inhomogeneities were adopted to evaluate the outcome of the proposed method; the MSRBDT algorithm was used to identify the gray matter (GM), white matter (WM), and cerebral-spinal fluid (CSF) in the brain tissues. The accuracy rates of GM, WM, and CSF segmentation, with spatial features (G, x, y, r, θ), were respectively greater than 0.9805, 0.9817, and 0.9871. In addition, images segmented with the MSRBDT algorithm were better than those obtained with the expectation maximization (EM) algorithm; brain tissue segmentation in MR images was significantly more precise. The proposed MSRBDT algorithm could be beneficial in clinical image segmentation.

AB - The purpose of this study was to improve the accuracy of tissue segmentation on brain magnetic resonance (MR) images preprocessed by multiscale retinex (MSR), segmented with a combined boosted decision tree (BDT) and MSR algorithm (hereinafter referred to as the MSRBDT algorithm). Simulated brain MR (SBMR) T1-weighted images of different noise levels and RF inhomogeneities were adopted to evaluate the outcome of the proposed method; the MSRBDT algorithm was used to identify the gray matter (GM), white matter (WM), and cerebral-spinal fluid (CSF) in the brain tissues. The accuracy rates of GM, WM, and CSF segmentation, with spatial features (G, x, y, r, θ), were respectively greater than 0.9805, 0.9817, and 0.9871. In addition, images segmented with the MSRBDT algorithm were better than those obtained with the expectation maximization (EM) algorithm; brain tissue segmentation in MR images was significantly more precise. The proposed MSRBDT algorithm could be beneficial in clinical image segmentation.

KW - Boosted decision tree

KW - Brain tissue

KW - Multiscale retinex

KW - Segmentation

KW - Spatial feature

UR - http://www.scopus.com/inward/record.url?scp=84857356581&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84857356581&partnerID=8YFLogxK

U2 - 10.1016/j.bspc.2011.04.001

DO - 10.1016/j.bspc.2011.04.001

M3 - Article

AN - SCOPUS:84857356581

VL - 7

SP - 129

EP - 140

JO - Biomedical Signal Processing and Control

JF - Biomedical Signal Processing and Control

SN - 1746-8094

IS - 2

ER -