Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach

Norhashimah Mohd Saad, S. A R Abu-Bakar, Sobri Muda, Musa Mokji

Research output: Chapter in Book/Report/Conference proceedingConference contribution

9 Citations (Scopus)

Abstract

This paper presents a segmentation of brain lesion from diffusion-weighted magnetic resonance images (DW-MRI or DWI) using a split and merge approach. The lesions are hyperintense lesion from tumour, acute infarction, haemorrhage and abscess, and hypointense lesion from chronic infarction and haemorrhage. Pre-processing is applied to the DWI for intensity normalization, background removal and intensity enhancement. Then, the split and merge algorithm is designed to segment the lesion. Histogram thresholding technique is used at each split level to detect pixels with either hyperintense or hypointense. Several statistical features are discussed and evaluated to select the best feature as homogeneity criteria. The analysis shows that mean and number of lesion pixels are the best homogeneity criteria. Hyperintense and hypointense lesions are segmented automatically by merging the regions that are homogenous according to the criteria.

Original languageEnglish
Title of host publicationProceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010
Pages475-480
Number of pages6
DOIs
Publication statusPublished - 2010
Event2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010 - Kuala Lumpur
Duration: 30 Nov 20102 Dec 2010

Other

Other2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010
CityKuala Lumpur
Period30/11/102/12/10

Fingerprint

Magnetic resonance imaging
Brain
Pixels
Magnetic resonance
Merging
Tumors
Processing

Keywords

  • brain lesion
  • DWI
  • segmentation
  • split and merge

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Saad, N. M., Abu-Bakar, S. A. R., Muda, S., & Mokji, M. (2010). Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach. In Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010 (pp. 475-480). [5742284] https://doi.org/10.1109/IECBES.2010.5742284

Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach. / Saad, Norhashimah Mohd; Abu-Bakar, S. A R; Muda, Sobri; Mokji, Musa.

Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010. 2010. p. 475-480 5742284.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Saad, NM, Abu-Bakar, SAR, Muda, S & Mokji, M 2010, Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach. in Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010., 5742284, pp. 475-480, 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010, Kuala Lumpur, 30/11/10. https://doi.org/10.1109/IECBES.2010.5742284
Saad NM, Abu-Bakar SAR, Muda S, Mokji M. Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach. In Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010. 2010. p. 475-480. 5742284 https://doi.org/10.1109/IECBES.2010.5742284
Saad, Norhashimah Mohd ; Abu-Bakar, S. A R ; Muda, Sobri ; Mokji, Musa. / Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach. Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010. 2010. pp. 475-480
@inproceedings{7909c9048e8541fd9d46ee3666a14974,
title = "Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach",
abstract = "This paper presents a segmentation of brain lesion from diffusion-weighted magnetic resonance images (DW-MRI or DWI) using a split and merge approach. The lesions are hyperintense lesion from tumour, acute infarction, haemorrhage and abscess, and hypointense lesion from chronic infarction and haemorrhage. Pre-processing is applied to the DWI for intensity normalization, background removal and intensity enhancement. Then, the split and merge algorithm is designed to segment the lesion. Histogram thresholding technique is used at each split level to detect pixels with either hyperintense or hypointense. Several statistical features are discussed and evaluated to select the best feature as homogeneity criteria. The analysis shows that mean and number of lesion pixels are the best homogeneity criteria. Hyperintense and hypointense lesions are segmented automatically by merging the regions that are homogenous according to the criteria.",
keywords = "brain lesion, DWI, segmentation, split and merge",
author = "Saad, {Norhashimah Mohd} and Abu-Bakar, {S. A R} and Sobri Muda and Musa Mokji",
year = "2010",
doi = "10.1109/IECBES.2010.5742284",
language = "English",
isbn = "9781424476008",
pages = "475--480",
booktitle = "Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010",

}

TY - GEN

T1 - Automated segmentation of brain lesion based on diffusion-weighted MRI using a split and merge approach

AU - Saad, Norhashimah Mohd

AU - Abu-Bakar, S. A R

AU - Muda, Sobri

AU - Mokji, Musa

PY - 2010

Y1 - 2010

N2 - This paper presents a segmentation of brain lesion from diffusion-weighted magnetic resonance images (DW-MRI or DWI) using a split and merge approach. The lesions are hyperintense lesion from tumour, acute infarction, haemorrhage and abscess, and hypointense lesion from chronic infarction and haemorrhage. Pre-processing is applied to the DWI for intensity normalization, background removal and intensity enhancement. Then, the split and merge algorithm is designed to segment the lesion. Histogram thresholding technique is used at each split level to detect pixels with either hyperintense or hypointense. Several statistical features are discussed and evaluated to select the best feature as homogeneity criteria. The analysis shows that mean and number of lesion pixels are the best homogeneity criteria. Hyperintense and hypointense lesions are segmented automatically by merging the regions that are homogenous according to the criteria.

AB - This paper presents a segmentation of brain lesion from diffusion-weighted magnetic resonance images (DW-MRI or DWI) using a split and merge approach. The lesions are hyperintense lesion from tumour, acute infarction, haemorrhage and abscess, and hypointense lesion from chronic infarction and haemorrhage. Pre-processing is applied to the DWI for intensity normalization, background removal and intensity enhancement. Then, the split and merge algorithm is designed to segment the lesion. Histogram thresholding technique is used at each split level to detect pixels with either hyperintense or hypointense. Several statistical features are discussed and evaluated to select the best feature as homogeneity criteria. The analysis shows that mean and number of lesion pixels are the best homogeneity criteria. Hyperintense and hypointense lesions are segmented automatically by merging the regions that are homogenous according to the criteria.

KW - brain lesion

KW - DWI

KW - segmentation

KW - split and merge

UR - http://www.scopus.com/inward/record.url?scp=79955413433&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79955413433&partnerID=8YFLogxK

U2 - 10.1109/IECBES.2010.5742284

DO - 10.1109/IECBES.2010.5742284

M3 - Conference contribution

AN - SCOPUS:79955413433

SN - 9781424476008

SP - 475

EP - 480

BT - Proceedings of 2010 IEEE EMBS Conference on Biomedical Engineering and Sciences, IECBES 2010

ER -