Forskning
Udskriv Udskriv
Switch language
Region Hovedstaden - en del af Københavns Universitetshospital
Udgivet

Reduced accuracy of MRI deep grey matter segmentation in multiple sclerosis: an evaluation of four automated methods against manual reference segmentations in a multi-center cohort

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

Harvard

APA

CBE

MLA

Vancouver

Author

Bibtex

@article{6fefe07236ff4d049642c073da69b14f,
title = "Reduced accuracy of MRI deep grey matter segmentation in multiple sclerosis: an evaluation of four automated methods against manual reference segmentations in a multi-center cohort",
abstract = "BACKGROUND: Deep grey matter (DGM) atrophy in multiple sclerosis (MS) and its relation to cognitive and clinical decline requires accurate measurements. MS pathology may deteriorate the performance of automated segmentation methods. Accuracy of DGM segmentation methods is compared between MS and controls, and the relation of performance with lesions and atrophy is studied.METHODS: On images of 21 MS subjects and 11 controls, three raters manually outlined caudate nucleus, putamen and thalamus; outlines were combined by majority voting. FSL-FIRST, FreeSurfer, Geodesic Information Flow and volBrain were evaluated. Performance was evaluated volumetrically (intra-class correlation coefficient (ICC)) and spatially (Dice similarity coefficient (DSC)). Spearman's correlations of DSC with global and local lesion volume, structure of interest volume (ROIV), and normalized brain volume (NBV) were assessed.RESULTS: ICC with manual volumes was mostly good and spatial agreement was high. MS exhibited significantly lower DSC than controls for thalamus and putamen. For some combinations of structure and method, DSC correlated negatively with lesion volume or positively with NBV or ROIV. Lesion-filling did not substantially change segmentations.CONCLUSIONS: Automated methods have impaired performance in patients. Performance generally deteriorated with higher lesion volume and lower NBV and ROIV, suggesting that these may contribute to the impaired performance.",
author = "{de Sitter}, Alexandra and Tom Verhoeven and Jessica Burggraaff and Yaou Liu and Jorge Simoes and Serena Ruggieri and Miklos Palotai and Iman Brouwer and Adriaan Versteeg and Viktor Wottschel and Stefan Ropele and Rocca, {Mara A} and Claudio Gasperini and Antonio Gallo and Yiannakas, {Marios C} and Alex Rovira and Christian Enzinger and Massimo Filippi and {De Stefano}, Nicola and Ludwig Kappos and Frederiksen, {Jette L} and Uitdehaag, {Bernard M J} and Frederik Barkhof and Guttmann, {Charles R G} and Hugo Vrenken and {MAGNIMS Study Group}",
year = "2020",
month = dec,
doi = "10.1007/s00415-020-10023-1",
language = "English",
volume = "267",
pages = "3541--3554",
journal = "Journal of Neurology",
issn = "0340-5354",
publisher = "Dr. Dietrich/Steinkopff Verlag",
number = "12",

}

RIS

TY - JOUR

T1 - Reduced accuracy of MRI deep grey matter segmentation in multiple sclerosis

T2 - an evaluation of four automated methods against manual reference segmentations in a multi-center cohort

AU - de Sitter, Alexandra

AU - Verhoeven, Tom

AU - Burggraaff, Jessica

AU - Liu, Yaou

AU - Simoes, Jorge

AU - Ruggieri, Serena

AU - Palotai, Miklos

AU - Brouwer, Iman

AU - Versteeg, Adriaan

AU - Wottschel, Viktor

AU - Ropele, Stefan

AU - Rocca, Mara A

AU - Gasperini, Claudio

AU - Gallo, Antonio

AU - Yiannakas, Marios C

AU - Rovira, Alex

AU - Enzinger, Christian

AU - Filippi, Massimo

AU - De Stefano, Nicola

AU - Kappos, Ludwig

AU - Frederiksen, Jette L

AU - Uitdehaag, Bernard M J

AU - Barkhof, Frederik

AU - Guttmann, Charles R G

AU - Vrenken, Hugo

AU - MAGNIMS Study Group

PY - 2020/12

Y1 - 2020/12

N2 - BACKGROUND: Deep grey matter (DGM) atrophy in multiple sclerosis (MS) and its relation to cognitive and clinical decline requires accurate measurements. MS pathology may deteriorate the performance of automated segmentation methods. Accuracy of DGM segmentation methods is compared between MS and controls, and the relation of performance with lesions and atrophy is studied.METHODS: On images of 21 MS subjects and 11 controls, three raters manually outlined caudate nucleus, putamen and thalamus; outlines were combined by majority voting. FSL-FIRST, FreeSurfer, Geodesic Information Flow and volBrain were evaluated. Performance was evaluated volumetrically (intra-class correlation coefficient (ICC)) and spatially (Dice similarity coefficient (DSC)). Spearman's correlations of DSC with global and local lesion volume, structure of interest volume (ROIV), and normalized brain volume (NBV) were assessed.RESULTS: ICC with manual volumes was mostly good and spatial agreement was high. MS exhibited significantly lower DSC than controls for thalamus and putamen. For some combinations of structure and method, DSC correlated negatively with lesion volume or positively with NBV or ROIV. Lesion-filling did not substantially change segmentations.CONCLUSIONS: Automated methods have impaired performance in patients. Performance generally deteriorated with higher lesion volume and lower NBV and ROIV, suggesting that these may contribute to the impaired performance.

AB - BACKGROUND: Deep grey matter (DGM) atrophy in multiple sclerosis (MS) and its relation to cognitive and clinical decline requires accurate measurements. MS pathology may deteriorate the performance of automated segmentation methods. Accuracy of DGM segmentation methods is compared between MS and controls, and the relation of performance with lesions and atrophy is studied.METHODS: On images of 21 MS subjects and 11 controls, three raters manually outlined caudate nucleus, putamen and thalamus; outlines were combined by majority voting. FSL-FIRST, FreeSurfer, Geodesic Information Flow and volBrain were evaluated. Performance was evaluated volumetrically (intra-class correlation coefficient (ICC)) and spatially (Dice similarity coefficient (DSC)). Spearman's correlations of DSC with global and local lesion volume, structure of interest volume (ROIV), and normalized brain volume (NBV) were assessed.RESULTS: ICC with manual volumes was mostly good and spatial agreement was high. MS exhibited significantly lower DSC than controls for thalamus and putamen. For some combinations of structure and method, DSC correlated negatively with lesion volume or positively with NBV or ROIV. Lesion-filling did not substantially change segmentations.CONCLUSIONS: Automated methods have impaired performance in patients. Performance generally deteriorated with higher lesion volume and lower NBV and ROIV, suggesting that these may contribute to the impaired performance.

U2 - 10.1007/s00415-020-10023-1

DO - 10.1007/s00415-020-10023-1

M3 - Journal article

C2 - 32621103

VL - 267

SP - 3541

EP - 3554

JO - Journal of Neurology

JF - Journal of Neurology

SN - 0340-5354

IS - 12

ER -

ID: 61516285