Home > Publications database > Deep learning-based 3D reconstruction of dentate nuclei in Friedreich’s ataxia from T2*weighted MR images > print

001     1052293
005     20260122203309.0
024 7 _ |a 10.1016/j.mlwa.2025.100790
|2 doi
024 7 _ |a 10.34734/FZJ-2026-00910
|2 datacite_doi
037 _ _ |a FZJ-2026-00910
082 _ _ |a 004
100 1 _ |a Sardhara, Trushal
|0 0000-0001-9915-3039
|b 0
245 _ _ |a Deep learning-based 3D reconstruction of dentate nuclei in Friedreich’s ataxia from T2*weighted MR images
260 _ _ |a Amsterdam
|c 2025
|b Elsevier
336 7 _ |a article
|2 DRIVER
336 7 _ |a Output Types/Journal article
|2 DataCite
336 7 _ |a Journal Article
|b journal
|m journal
|0 PUB:(DE-HGF)16
|s 1769088056_20964
|2 PUB:(DE-HGF)
336 7 _ |a ARTICLE
|2 BibTeX
336 7 _ |a JOURNAL_ARTICLE
|2 ORCID
336 7 _ |a Journal Article
|0 0
|2 EndNote
520 _ _ |a Dentate nucleus (DN) degeneration is a key neuropathological feature in Friedreich’s ataxia (FRDA), and its accurate quantification is critical for understanding disease progression. However, its visualization and volumetry require iron-sensitive imaging techniques and time-consuming segmentation procedures, posing challenges for conventional ML approaches due to small datasets typical of rare diseases. We present a transfer learning–based machine learning pipeline for automated DN segmentation that directly uses standard T2*-weighted Magnetic Resonance Imaging (MRI), which highlights the DN without additional processing, and is designed to perform robustly with limited annotated data. Using 38 manually labeled subjects (18 FRDA, 20 controls), the model was validated via five-fold cross-validation and an independent hold-out test set, achieving Dice scores of 0.81–0.87 and outperforming classical atlas-based methods. Pretraining improved performance by ∼10% in patients and >5% in controls. Applied to 181 longitudinal scans from 33 FRDA patients and 33 controls, the model revealed significantly reduced DN volumes in FRDA, with reductions correlating with disease duration and clinical severity over time. Our approach provides a scalable and reproducible segmentation framework, requiring minimal annotated data and no preprocessing, while demonstrating robust performance across cross-validation and independent testing. Additionally, it enables the first longitudinal volumetric analysis of DN in FRDA using standard T2*-weighted MRI, demonstrating its practical utility for monitoring neurodegenerative changes. Overall, this work illustrates how transfer learning can overcome data scarcity in rare diseases and provides a robust methodology for automated MRI segmentation in both research and clinical applications.
536 _ _ |a 5251 - Multilevel Brain Organization and Variability (POF4-525)
|0 G:(DE-HGF)POF4-5251
|c POF4-525
|f POF IV
|x 0
588 _ _ |a Dataset connected to CrossRef, Journals: juser.fz-juelich.de
700 1 _ |a Dadsena, Ravi
|0 P:(DE-HGF)0
|b 1
700 1 _ |a Aydin, Roland C.
|0 P:(DE-HGF)0
|b 2
700 1 _ |a Hilgers, Ralf-Dieter
|b 3
700 1 _ |a Horn, Leon
|b 4
700 1 _ |a Schulz, Jörg B.
|0 P:(DE-Juel1)171786
|b 5
700 1 _ |a Reetz, Kathrin
|0 P:(DE-Juel1)177889
|b 6
|e Corresponding author
700 1 _ |a Romanzetti, Sandro
|b 7
700 1 _ |a Dogan, Imis
|b 8
700 1 _ |a Lischewski, Stella A.
|b 9
700 1 _ |a Konrad, Kerstin
|0 P:(DE-Juel1)174172
|b 10
700 1 _ |a Pishnamaz, Miguel
|b 11
700 1 _ |a Praster, Maximillian
|b 12
700 1 _ |a Clavel, Thomas
|b 13
700 1 _ |a Jankowski, Vera
|b 14
700 1 _ |a Jankowski, Joachim
|b 15
700 1 _ |a Pabst, Oliver
|b 16
700 1 _ |a Marx-Schütt, Katharina
|b 17
700 1 _ |a Marx, Nikolaus
|b 18
700 1 _ |a Möllmann, Julia
|b 19
700 1 _ |a Jacobsen, Malte
|b 20
700 1 _ |a Dukart, Juergen
|0 P:(DE-Juel1)177772
|b 21
700 1 _ |a Eickhoff, Simon
|0 P:(DE-Juel1)131678
|b 22
773 _ _ |a 10.1016/j.mlwa.2025.100790
|g Vol. 22, p. 100790 -
|0 PERI:(DE-600)3052291-2
|p 100790 -
|t Machine learning with applications
|v 22
|y 2025
|x 2666-8270
856 4 _ |u https://juser.fz-juelich.de/record/1052293/files/1-s2.0-S2666827025001732-main.pdf
|y OpenAccess
909 C O |o oai:juser.fz-juelich.de:1052293
|p openaire
|p open_access
|p VDB
|p driver
|p dnbdelivery
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 5
|6 P:(DE-Juel1)171786
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 6
|6 P:(DE-Juel1)177889
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 10
|6 P:(DE-Juel1)174172
910 1 _ |a Forschungszentrum Jülich
|0 I:(DE-588b)5008462-8
|k FZJ
|b 22
|6 P:(DE-Juel1)131678
913 1 _ |a DE-HGF
|b Key Technologies
|l Natural, Artificial and Cognitive Information Processing
|1 G:(DE-HGF)POF4-520
|0 G:(DE-HGF)POF4-525
|3 G:(DE-HGF)POF4
|2 G:(DE-HGF)POF4-500
|4 G:(DE-HGF)POF
|v Decoding Brain Organization and Dysfunction
|9 G:(DE-HGF)POF4-5251
|x 0
915 _ _ |a DBCoverage
|0 StatID:(DE-HGF)0150
|2 StatID
|b Web of Science Core Collection
|d 2025-01-07
915 _ _ |a Creative Commons Attribution CC BY 4.0
|0 LIC:(DE-HGF)CCBY4
|2 HGFVOC
915 _ _ |a WoS
|0 StatID:(DE-HGF)0112
|2 StatID
|b Emerging Sources Citation Index
|d 2025-01-07
915 _ _ |a DBCoverage
|0 StatID:(DE-HGF)0501
|2 StatID
|b DOAJ Seal
|d 2021-01-29T11:25:19Z
915 _ _ |a DBCoverage
|0 StatID:(DE-HGF)0500
|2 StatID
|b DOAJ
|d 2021-01-29T11:25:19Z
915 _ _ |a Fees
|0 StatID:(DE-HGF)0700
|2 StatID
|d 2025-01-07
915 _ _ |a OpenAccess
|0 StatID:(DE-HGF)0510
|2 StatID
915 _ _ |a Peer Review
|0 StatID:(DE-HGF)0030
|2 StatID
|b DOAJ : Anonymous peer review
|d 2021-01-29T11:25:19Z
915 _ _ |a Article Processing Charges
|0 StatID:(DE-HGF)0561
|2 StatID
|d 2025-01-07
915 _ _ |a DBCoverage
|0 StatID:(DE-HGF)0300
|2 StatID
|b Medline
|d 2025-01-07
915 _ _ |a DBCoverage
|0 StatID:(DE-HGF)0199
|2 StatID
|b Clarivate Analytics Master Journal List
|d 2025-01-07
920 _ _ |l yes
920 1 _ |0 I:(DE-Juel1)INM-7-20090406
|k INM-7
|l Gehirn & Verhalten
|x 0
980 _ _ |a journal
980 _ _ |a VDB
980 _ _ |a UNRESTRICTED
980 _ _ |a I:(DE-Juel1)INM-7-20090406
980 1 _ |a FullTexts

LibraryCollectionCLSMajorCLSMinorLanguageAuthor
Marc 21