Automated detection of multiple sclerosis candidate regions in MR images: False-positive removal with use of an ANN-controlled level-set method

Jumpei Kuwazuru, Hidetaka Arimura, Shingo Kakeda, Daisuke Yamamoto, Taiki Magome, Yasuo Yamashita, Masafumi Ohki, Fukai Toyofuku, Yukunori Korogi

Research output: Contribution to journalArticle

4 Citations (Scopus)

Abstract

Our purpose in this study was to develop an automated segmentation scheme for multiple sclerosis (MS) lesions in magnetic resonance images using an artificial neural network (ANN)-controlled level-set method. Forty-nine slices with T1-weighted, T2-weighted, and fluid-attenuated inversion recovery images were selected from six examinations of three MS patients including 168 MS lesions for this study. First, MS lesions were enhanced by background subtraction. Initial regions of MS candidates were detected based on a multiple-gray-level thresholding technique and a region-growing technique on the subtraction image. Then, final regions of MS candidates were determined by application of a proposed segmentation method using an ANN-controlled level-set method, which was used for reduction of false positives (FPs) as well as more accurate segmentation. Finally, all candidate regions were classified into true positive and FP candidate regions by use of a support vector machine. As the result of a leave-one-candidate-out test method, the detection sensitivity for MS lesions increased from 64.9 to 75.0% while decreasing the number of FPs per slice from 19.9 to 4.4 compared with a previous study. The proposed scheme improved the sensitivity and the number of FPs in the detection of MS lesions.

Original languageEnglish
Pages (from-to)105-113
Number of pages9
JournalRadiological physics and technology
Volume5
Issue number1
DOIs
Publication statusPublished - Jan 1 2012

Fingerprint

lesions
Multiple Sclerosis
subtraction
Subtraction Technique
magnetic resonance
examination
recovery
inversions
fluids
sensitivity
Magnetic Resonance Spectroscopy

All Science Journal Classification (ASJC) codes

  • Radiation
  • Physical Therapy, Sports Therapy and Rehabilitation
  • Radiology Nuclear Medicine and imaging

Cite this

Automated detection of multiple sclerosis candidate regions in MR images : False-positive removal with use of an ANN-controlled level-set method. / Kuwazuru, Jumpei; Arimura, Hidetaka; Kakeda, Shingo; Yamamoto, Daisuke; Magome, Taiki; Yamashita, Yasuo; Ohki, Masafumi; Toyofuku, Fukai; Korogi, Yukunori.

In: Radiological physics and technology, Vol. 5, No. 1, 01.01.2012, p. 105-113.

Research output: Contribution to journalArticle

Kuwazuru, Jumpei ; Arimura, Hidetaka ; Kakeda, Shingo ; Yamamoto, Daisuke ; Magome, Taiki ; Yamashita, Yasuo ; Ohki, Masafumi ; Toyofuku, Fukai ; Korogi, Yukunori. / Automated detection of multiple sclerosis candidate regions in MR images : False-positive removal with use of an ANN-controlled level-set method. In: Radiological physics and technology. 2012 ; Vol. 5, No. 1. pp. 105-113.
@article{aff02c3c0a314c00b9f9ea110e785aa8,
title = "Automated detection of multiple sclerosis candidate regions in MR images: False-positive removal with use of an ANN-controlled level-set method",
abstract = "Our purpose in this study was to develop an automated segmentation scheme for multiple sclerosis (MS) lesions in magnetic resonance images using an artificial neural network (ANN)-controlled level-set method. Forty-nine slices with T1-weighted, T2-weighted, and fluid-attenuated inversion recovery images were selected from six examinations of three MS patients including 168 MS lesions for this study. First, MS lesions were enhanced by background subtraction. Initial regions of MS candidates were detected based on a multiple-gray-level thresholding technique and a region-growing technique on the subtraction image. Then, final regions of MS candidates were determined by application of a proposed segmentation method using an ANN-controlled level-set method, which was used for reduction of false positives (FPs) as well as more accurate segmentation. Finally, all candidate regions were classified into true positive and FP candidate regions by use of a support vector machine. As the result of a leave-one-candidate-out test method, the detection sensitivity for MS lesions increased from 64.9 to 75.0{\%} while decreasing the number of FPs per slice from 19.9 to 4.4 compared with a previous study. The proposed scheme improved the sensitivity and the number of FPs in the detection of MS lesions.",
author = "Jumpei Kuwazuru and Hidetaka Arimura and Shingo Kakeda and Daisuke Yamamoto and Taiki Magome and Yasuo Yamashita and Masafumi Ohki and Fukai Toyofuku and Yukunori Korogi",
year = "2012",
month = "1",
day = "1",
doi = "10.1007/s12194-011-0141-2",
language = "English",
volume = "5",
pages = "105--113",
journal = "Radiological Physics and Technology",
issn = "1865-0333",
publisher = "Springer Japan",
number = "1",

}

TY - JOUR

T1 - Automated detection of multiple sclerosis candidate regions in MR images

T2 - False-positive removal with use of an ANN-controlled level-set method

AU - Kuwazuru, Jumpei

AU - Arimura, Hidetaka

AU - Kakeda, Shingo

AU - Yamamoto, Daisuke

AU - Magome, Taiki

AU - Yamashita, Yasuo

AU - Ohki, Masafumi

AU - Toyofuku, Fukai

AU - Korogi, Yukunori

PY - 2012/1/1

Y1 - 2012/1/1

N2 - Our purpose in this study was to develop an automated segmentation scheme for multiple sclerosis (MS) lesions in magnetic resonance images using an artificial neural network (ANN)-controlled level-set method. Forty-nine slices with T1-weighted, T2-weighted, and fluid-attenuated inversion recovery images were selected from six examinations of three MS patients including 168 MS lesions for this study. First, MS lesions were enhanced by background subtraction. Initial regions of MS candidates were detected based on a multiple-gray-level thresholding technique and a region-growing technique on the subtraction image. Then, final regions of MS candidates were determined by application of a proposed segmentation method using an ANN-controlled level-set method, which was used for reduction of false positives (FPs) as well as more accurate segmentation. Finally, all candidate regions were classified into true positive and FP candidate regions by use of a support vector machine. As the result of a leave-one-candidate-out test method, the detection sensitivity for MS lesions increased from 64.9 to 75.0% while decreasing the number of FPs per slice from 19.9 to 4.4 compared with a previous study. The proposed scheme improved the sensitivity and the number of FPs in the detection of MS lesions.

AB - Our purpose in this study was to develop an automated segmentation scheme for multiple sclerosis (MS) lesions in magnetic resonance images using an artificial neural network (ANN)-controlled level-set method. Forty-nine slices with T1-weighted, T2-weighted, and fluid-attenuated inversion recovery images were selected from six examinations of three MS patients including 168 MS lesions for this study. First, MS lesions were enhanced by background subtraction. Initial regions of MS candidates were detected based on a multiple-gray-level thresholding technique and a region-growing technique on the subtraction image. Then, final regions of MS candidates were determined by application of a proposed segmentation method using an ANN-controlled level-set method, which was used for reduction of false positives (FPs) as well as more accurate segmentation. Finally, all candidate regions were classified into true positive and FP candidate regions by use of a support vector machine. As the result of a leave-one-candidate-out test method, the detection sensitivity for MS lesions increased from 64.9 to 75.0% while decreasing the number of FPs per slice from 19.9 to 4.4 compared with a previous study. The proposed scheme improved the sensitivity and the number of FPs in the detection of MS lesions.

UR - http://www.scopus.com/inward/record.url?scp=84858867290&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84858867290&partnerID=8YFLogxK

U2 - 10.1007/s12194-011-0141-2

DO - 10.1007/s12194-011-0141-2

M3 - Article

C2 - 22139608

AN - SCOPUS:84858867290

VL - 5

SP - 105

EP - 113

JO - Radiological Physics and Technology

JF - Radiological Physics and Technology

SN - 1865-0333

IS - 1

ER -