Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Automated Hippocampal Subfield Segmentation at 7T MRI

Wisse, L E M LU orcid ; Kuijf, H J ; Honingh, A M ; Wang, H ; Pluta, J B ; Das, S R ; Wolk, D A ; Zwanenburg, J J M ; Yushkevich, P A and Geerlings, M I (2016) In AJNR. American journal of neuroradiology 37(6). p.7-1050
Abstract

BACKGROUND AND PURPOSE: High resolution 7T MRI is increasingly used to investigate hippocampal subfields in vivo, but most studies rely on manual segmentation which is labor intensive. We aimed to evaluate an automated technique to segment hippocampal subfields and the entorhinal cortex at 7T MRI.

MATERIALS AND METHODS: The cornu ammonis (CA)1, CA2, CA3, dentate gyrus, subiculum, and entorhinal cortex were manually segmented, covering most of the long axis of the hippocampus on 0.70-mm(3) T2-weighted 7T images of 26 participants (59 ± 9 years, 46% men). The automated segmentation of hippocampal subfields approach was applied and evaluated by using leave-one-out cross-validation.

RESULTS: Comparison of automated segmentations... (More)

BACKGROUND AND PURPOSE: High resolution 7T MRI is increasingly used to investigate hippocampal subfields in vivo, but most studies rely on manual segmentation which is labor intensive. We aimed to evaluate an automated technique to segment hippocampal subfields and the entorhinal cortex at 7T MRI.

MATERIALS AND METHODS: The cornu ammonis (CA)1, CA2, CA3, dentate gyrus, subiculum, and entorhinal cortex were manually segmented, covering most of the long axis of the hippocampus on 0.70-mm(3) T2-weighted 7T images of 26 participants (59 ± 9 years, 46% men). The automated segmentation of hippocampal subfields approach was applied and evaluated by using leave-one-out cross-validation.

RESULTS: Comparison of automated segmentations with corresponding manual segmentations yielded a Dice similarity coefficient of >0.75 for CA1, the dentate gyrus, subiculum, and entorhinal cortex and >0.54 for CA2 and CA3. Intraclass correlation coefficients were >0.74 for CA1, the dentate gyrus, and subiculum; and >0.43 for CA2, CA3, and the entorhinal cortex. Restricting the comparison of the entorhinal cortex segmentation to a smaller range along the anteroposterior axis improved both intraclass correlation coefficients (left: 0.71; right: 0.82) and Dice similarity coefficients (left: 0.78; right: 0.77). The accuracy of the automated segmentation versus a manual rater was lower, though only slightly for most subfields, than the intrarater reliability of an expert manual rater, but it was similar to or slightly higher than the accuracy of an expert-versus-manual rater with ∼170 hours of training for almost all subfields.

CONCLUSIONS: This work demonstrates the feasibility of using a computational technique to automatically label hippocampal subfields and the entorhinal cortex at 7T MRI, with a high accuracy for most subfields that is competitive with the labor-intensive manual segmentation. The software and atlas are publicly available: http://www.nitrc.org/projects/ashs/.

(Less)
Please use this url to cite or link to this publication:
author
; ; ; ; ; ; ; ; and
publishing date
type
Contribution to journal
publication status
published
keywords
Aged, Automation, CA1 Region, Hippocampal/diagnostic imaging, CA2 Region, Hippocampal/diagnostic imaging, CA3 Region, Hippocampal/diagnostic imaging, Dentate Gyrus/diagnostic imaging, Entorhinal Cortex/diagnostic imaging, Female, Hippocampus/diagnostic imaging, Humans, Image Processing, Computer-Assisted, Magnetic Resonance Imaging/methods, Male, Middle Aged, Observer Variation, Reproducibility of Results
in
AJNR. American journal of neuroradiology
volume
37
issue
6
pages
8 pages
publisher
American Society of Neuroradiology
external identifiers
  • scopus:84974625486
  • pmid:26846925
ISSN
1936-959X
DOI
10.3174/ajnr.A4659
language
English
LU publication?
no
additional info
© 2016 by American Journal of Neuroradiology.
id
d9084386-8c41-4f6a-a6d5-069e87d7224e
date added to LUP
2024-02-28 14:37:46
date last changed
2024-03-30 22:18:01
@article{d9084386-8c41-4f6a-a6d5-069e87d7224e,
  abstract     = {{<p>BACKGROUND AND PURPOSE: High resolution 7T MRI is increasingly used to investigate hippocampal subfields in vivo, but most studies rely on manual segmentation which is labor intensive. We aimed to evaluate an automated technique to segment hippocampal subfields and the entorhinal cortex at 7T MRI.</p><p>MATERIALS AND METHODS: The cornu ammonis (CA)1, CA2, CA3, dentate gyrus, subiculum, and entorhinal cortex were manually segmented, covering most of the long axis of the hippocampus on 0.70-mm(3) T2-weighted 7T images of 26 participants (59 ± 9 years, 46% men). The automated segmentation of hippocampal subfields approach was applied and evaluated by using leave-one-out cross-validation.</p><p>RESULTS: Comparison of automated segmentations with corresponding manual segmentations yielded a Dice similarity coefficient of &gt;0.75 for CA1, the dentate gyrus, subiculum, and entorhinal cortex and &gt;0.54 for CA2 and CA3. Intraclass correlation coefficients were &gt;0.74 for CA1, the dentate gyrus, and subiculum; and &gt;0.43 for CA2, CA3, and the entorhinal cortex. Restricting the comparison of the entorhinal cortex segmentation to a smaller range along the anteroposterior axis improved both intraclass correlation coefficients (left: 0.71; right: 0.82) and Dice similarity coefficients (left: 0.78; right: 0.77). The accuracy of the automated segmentation versus a manual rater was lower, though only slightly for most subfields, than the intrarater reliability of an expert manual rater, but it was similar to or slightly higher than the accuracy of an expert-versus-manual rater with ∼170 hours of training for almost all subfields.</p><p>CONCLUSIONS: This work demonstrates the feasibility of using a computational technique to automatically label hippocampal subfields and the entorhinal cortex at 7T MRI, with a high accuracy for most subfields that is competitive with the labor-intensive manual segmentation. The software and atlas are publicly available: http://www.nitrc.org/projects/ashs/.</p>}},
  author       = {{Wisse, L E M and Kuijf, H J and Honingh, A M and Wang, H and Pluta, J B and Das, S R and Wolk, D A and Zwanenburg, J J M and Yushkevich, P A and Geerlings, M I}},
  issn         = {{1936-959X}},
  keywords     = {{Aged; Automation; CA1 Region, Hippocampal/diagnostic imaging; CA2 Region, Hippocampal/diagnostic imaging; CA3 Region, Hippocampal/diagnostic imaging; Dentate Gyrus/diagnostic imaging; Entorhinal Cortex/diagnostic imaging; Female; Hippocampus/diagnostic imaging; Humans; Image Processing, Computer-Assisted; Magnetic Resonance Imaging/methods; Male; Middle Aged; Observer Variation; Reproducibility of Results}},
  language     = {{eng}},
  number       = {{6}},
  pages        = {{7--1050}},
  publisher    = {{American Society of Neuroradiology}},
  series       = {{AJNR. American journal of neuroradiology}},
  title        = {{Automated Hippocampal Subfield Segmentation at 7T MRI}},
  url          = {{http://dx.doi.org/10.3174/ajnr.A4659}},
  doi          = {{10.3174/ajnr.A4659}},
  volume       = {{37}},
  year         = {{2016}},
}