Fast and robust multi-atlas segmentation of brain magnetic resonance images

Jyrki Lötjönen (Corresponding Author), Robin Wolz, Juha Koikkalainen, Lennart Thurfjell, Gunhild Waldemar, Hilkka Soininen, Daniel Rueckert

Research output: Contribution to journalArticleScientificpeer-review

271 Citations (Scopus)

Abstract

We introduce an optimised pipeline for multi-atlas brain MRI segmentation. Both accuracy and speed of segmentation are considered. We study different similarity measures used in non-rigid registration. We show that intensity differences for intensity normalised images can be used instead of standard normalised mutual information in registration without compromising the accuracy but leading to threefold decrease in the computation time. We study and validate also different methods for atlas selection. Finally, we propose two new approaches for combining multi-atlas segmentation and intensity modelling based on segmentation using expectation maximisation (EM) and optimisation via graph cuts. The segmentation pipeline is evaluated with two data cohorts: IBSR data (N = 18, six subcortial structures: thalamus, caudate, putamen, pallidum, hippocampus, amygdala) and ADNI data (N =   60, hippocampus). The average similarity index between automatically and manually generated volumes was 0.849 (IBSR, six subcortical structures) and 0.880 (ADNI, hippocampus). The correlation coefficient for hippocampal volumes was 0.95 with the ADNI data. The computation time using a standard multicore PC computer was about 3–4 min. Our results compare favourably with other recently published results.
Original languageEnglish
Pages (from-to)2352-2365
JournalNeuroImage
Volume49
Issue number3
DOIs
Publication statusPublished - 2010
MoE publication typeA1 Journal article-refereed

Fingerprint

Atlases
Hippocampus
Magnetic Resonance Spectroscopy
Brain
Globus Pallidus
Putamen
Amygdala
Thalamus

Keywords

  • MRI
  • Segmentation
  • Atlases
  • Registration
  • Hippocampus

Cite this

Lötjönen, J., Wolz, R., Koikkalainen, J., Thurfjell, L., Waldemar, G., Soininen, H., & Rueckert, D. (2010). Fast and robust multi-atlas segmentation of brain magnetic resonance images. NeuroImage, 49(3), 2352-2365. https://doi.org/10.1016/j.neuroimage.2009.10.026
Lötjönen, Jyrki ; Wolz, Robin ; Koikkalainen, Juha ; Thurfjell, Lennart ; Waldemar, Gunhild ; Soininen, Hilkka ; Rueckert, Daniel. / Fast and robust multi-atlas segmentation of brain magnetic resonance images. In: NeuroImage. 2010 ; Vol. 49, No. 3. pp. 2352-2365.
@article{09109e2187a84351b003d319cfb9f8ca,
title = "Fast and robust multi-atlas segmentation of brain magnetic resonance images",
abstract = "We introduce an optimised pipeline for multi-atlas brain MRI segmentation. Both accuracy and speed of segmentation are considered. We study different similarity measures used in non-rigid registration. We show that intensity differences for intensity normalised images can be used instead of standard normalised mutual information in registration without compromising the accuracy but leading to threefold decrease in the computation time. We study and validate also different methods for atlas selection. Finally, we propose two new approaches for combining multi-atlas segmentation and intensity modelling based on segmentation using expectation maximisation (EM) and optimisation via graph cuts. The segmentation pipeline is evaluated with two data cohorts: IBSR data (N = 18, six subcortial structures: thalamus, caudate, putamen, pallidum, hippocampus, amygdala) and ADNI data (N =   60, hippocampus). The average similarity index between automatically and manually generated volumes was 0.849 (IBSR, six subcortical structures) and 0.880 (ADNI, hippocampus). The correlation coefficient for hippocampal volumes was 0.95 with the ADNI data. The computation time using a standard multicore PC computer was about 3–4 min. Our results compare favourably with other recently published results.",
keywords = "MRI, Segmentation, Atlases, Registration, Hippocampus",
author = "Jyrki L{\"o}tj{\"o}nen and Robin Wolz and Juha Koikkalainen and Lennart Thurfjell and Gunhild Waldemar and Hilkka Soininen and Daniel Rueckert",
year = "2010",
doi = "10.1016/j.neuroimage.2009.10.026",
language = "English",
volume = "49",
pages = "2352--2365",
journal = "NeuroImage",
issn = "1053-8119",
publisher = "Academic Press",
number = "3",

}

Lötjönen, J, Wolz, R, Koikkalainen, J, Thurfjell, L, Waldemar, G, Soininen, H & Rueckert, D 2010, 'Fast and robust multi-atlas segmentation of brain magnetic resonance images', NeuroImage, vol. 49, no. 3, pp. 2352-2365. https://doi.org/10.1016/j.neuroimage.2009.10.026

Fast and robust multi-atlas segmentation of brain magnetic resonance images. / Lötjönen, Jyrki (Corresponding Author); Wolz, Robin; Koikkalainen, Juha; Thurfjell, Lennart; Waldemar, Gunhild; Soininen, Hilkka; Rueckert, Daniel.

In: NeuroImage, Vol. 49, No. 3, 2010, p. 2352-2365.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Fast and robust multi-atlas segmentation of brain magnetic resonance images

AU - Lötjönen, Jyrki

AU - Wolz, Robin

AU - Koikkalainen, Juha

AU - Thurfjell, Lennart

AU - Waldemar, Gunhild

AU - Soininen, Hilkka

AU - Rueckert, Daniel

PY - 2010

Y1 - 2010

N2 - We introduce an optimised pipeline for multi-atlas brain MRI segmentation. Both accuracy and speed of segmentation are considered. We study different similarity measures used in non-rigid registration. We show that intensity differences for intensity normalised images can be used instead of standard normalised mutual information in registration without compromising the accuracy but leading to threefold decrease in the computation time. We study and validate also different methods for atlas selection. Finally, we propose two new approaches for combining multi-atlas segmentation and intensity modelling based on segmentation using expectation maximisation (EM) and optimisation via graph cuts. The segmentation pipeline is evaluated with two data cohorts: IBSR data (N = 18, six subcortial structures: thalamus, caudate, putamen, pallidum, hippocampus, amygdala) and ADNI data (N =   60, hippocampus). The average similarity index between automatically and manually generated volumes was 0.849 (IBSR, six subcortical structures) and 0.880 (ADNI, hippocampus). The correlation coefficient for hippocampal volumes was 0.95 with the ADNI data. The computation time using a standard multicore PC computer was about 3–4 min. Our results compare favourably with other recently published results.

AB - We introduce an optimised pipeline for multi-atlas brain MRI segmentation. Both accuracy and speed of segmentation are considered. We study different similarity measures used in non-rigid registration. We show that intensity differences for intensity normalised images can be used instead of standard normalised mutual information in registration without compromising the accuracy but leading to threefold decrease in the computation time. We study and validate also different methods for atlas selection. Finally, we propose two new approaches for combining multi-atlas segmentation and intensity modelling based on segmentation using expectation maximisation (EM) and optimisation via graph cuts. The segmentation pipeline is evaluated with two data cohorts: IBSR data (N = 18, six subcortial structures: thalamus, caudate, putamen, pallidum, hippocampus, amygdala) and ADNI data (N =   60, hippocampus). The average similarity index between automatically and manually generated volumes was 0.849 (IBSR, six subcortical structures) and 0.880 (ADNI, hippocampus). The correlation coefficient for hippocampal volumes was 0.95 with the ADNI data. The computation time using a standard multicore PC computer was about 3–4 min. Our results compare favourably with other recently published results.

KW - MRI

KW - Segmentation

KW - Atlases

KW - Registration

KW - Hippocampus

U2 - 10.1016/j.neuroimage.2009.10.026

DO - 10.1016/j.neuroimage.2009.10.026

M3 - Article

VL - 49

SP - 2352

EP - 2365

JO - NeuroImage

JF - NeuroImage

SN - 1053-8119

IS - 3

ER -

Lötjönen J, Wolz R, Koikkalainen J, Thurfjell L, Waldemar G, Soininen H et al. Fast and robust multi-atlas segmentation of brain magnetic resonance images. NeuroImage. 2010;49(3):2352-2365. https://doi.org/10.1016/j.neuroimage.2009.10.026