Repository logo

Infoscience

  • English
  • French
Log In
Logo EPFL, École polytechnique fédérale de Lausanne

Infoscience

  • English
  • French
Log In
  1. Home
  2. Academic and Research Output
  3. Conferences, Workshops, Symposiums, and Seminars
  4. Ground-Truth Effects in Learning-Based Fiber Orientation Distribution Estimation in Neonatal Brains
 
conference paper

Ground-Truth Effects in Learning-Based Fiber Orientation Distribution Estimation in Neonatal Brains

Lin, Rizhong  orcid-logo
•
Kebiri, Hamza  
•
Gholipour, Ali
Show more
Chamberland, Maxime
•
Hendriks, Tom
Show more
April 18, 2025
Computational Diffusion MRI - 15th International Workshop, CDMRI 2024, Held in Conjunction with MICCAI 2024. Proceedings
15th International Workshop, CDMRI 2024, Held in Conjunction with MICCAI 2024

Diffusion Magnetic Resonance Imaging (dMRI) is a non-invasive method for depicting brain microstructure in vivo. Fiber orientation distributions (FODs) are mathematical representations extensively used to map white matter fiber configurations. Recently, FOD estimation with deep neural networks has seen growing success, in particular, those of neonates estimated with fewer diffusion measurements. These methods are mostly trained on target FODs reconstructed with multi-shell multi-tissue constrained spherical deconvolution (MSMT-CSD), which might not be the ideal ground truth for developing brains. Here, we investigate this hypothesis by training a state-of-the-art model based on the U-Net architecture on both MSMT-CSD and single-shell three-tissue constrained spherical deconvolution (SS3T-CSD). Our results suggest that SS3T-CSD might be more suited for neonatal brains, given that the ratio between single and multiple fiber-estimated voxels with SS3T-CSD is more realistic compared to MSMT-CSD. Additionally, increasing the number of input gradient directions significantly improves performance with SS3T-CSD over MSMT-CSD. Finally, in an age domain-shift setting, SS3T-CSD maintains robust performance across age groups, indicating its potential for more accurate neonatal brain imaging.

  • Details
  • Metrics
Type
conference paper
DOI
10.1007/978-3-031-86920-4_3
Scopus ID

2-s2.0-105003627857

Author(s)
Lin, Rizhong  orcid-logo

École Polytechnique Fédérale de Lausanne

Kebiri, Hamza  

École Polytechnique Fédérale de Lausanne

Gholipour, Ali

Boston Children's Hospital

Chen, Yufei

Tongji University

Thiran, Jean Philippe  

École Polytechnique Fédérale de Lausanne

Karimi, Davood

Boston Children's Hospital

Bach Cuadra, Meritxell  

École Polytechnique Fédérale de Lausanne

Editors
Chamberland, Maxime
•
Hendriks, Tom
•
Karaman, Muge
•
Mito, Remika
•
Newlin, Nancy
•
Thompson, Elinor
Date Issued

2025-04-18

Publisher

Springer Science and Business Media Deutschland GmbH

Published in
Computational Diffusion MRI - 15th International Workshop, CDMRI 2024, Held in Conjunction with MICCAI 2024. Proceedings
ISBN of the book

978-3-031-86920-4

Series title/Series vol.

Lecture Notes in Computer Science; 15171

ISSN (of the series)

1611-3349

0302-9743

Start page

24

End page

34

Subjects

Age domain shift

•

Deep learning

•

FOD estimation

•

MSMT-CSD

•

Neonatal brain

•

SS3T-CSD

Editorial or Peer reviewed

REVIEWED

Written at

EPFL

EPFL units
MLO  
MIPLAB  
LTS5  
Event nameEvent acronymEvent placeEvent date
15th International Workshop, CDMRI 2024, Held in Conjunction with MICCAI 2024

Marrakesh, Morocco

2024-10-06 - 2024-10-06

FunderFunding(s)Grant NumberGrant URL

Leenaards and Jeantet Foundations

University of Lausanne

University of Geneva

Show more
Available on Infoscience
May 12, 2025
Use this identifier to reference this record
https://infoscience.epfl.ch/handle/20.500.14299/250007
Logo EPFL, École polytechnique fédérale de Lausanne
  • Contact
  • infoscience@epfl.ch

  • Follow us on Facebook
  • Follow us on Instagram
  • Follow us on LinkedIn
  • Follow us on X
  • Follow us on Youtube
AccessibilityLegal noticePrivacy policyCookie settingsEnd User AgreementGet helpFeedback

Infoscience is a service managed and provided by the Library and IT Services of EPFL. © EPFL, tous droits réservés