EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Automated Segmentation of Median Nerve Cross‐Sectional Area in Healthy Controls and Patients With Carpal Tunnel Syndrome Using a Convolutional Neural Network

Tom B G Olde Dubbelink iD, Erik Willemen iD, Bas Schippers, Jan Meulstee, Frouke A P Nijhuis iD, Ronald H M A Bartels iD, Wim I M Verhagen iD

Muscle & Nerve · 2026

DOI: 10.1002/mus.70167 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Muscle & Nerve, 2026. Verfasst von Tom B G Olde Dubbelink und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Neurology, Canisius Wilhelmina Hospital, Nijmegen, the Netherlands. Hauptthemen laut MEDLINE: Carpal Tunnel Syndrome, Median Nerve, Neural Networks, Computer. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: Canisius Wilhelmina Hospital Science Fund.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Introduction/Aims The cross‐sectional area (CSA) of the median nerve (MN) is a key parameter for confirming carpal tunnel syndrome (CTS) with ultrasound. This study evaluates the performance of a convolutional neural network (CNN) with a 2D U‐Net architecture for automated MN CSA segmentation in both healthy individuals and patients with CTS. Automated segmentation supports large‐scale analysis, improves standardization across centers, and may serve as a foundation for future diagnostic tools.

Methods

Three hundred static ultrasound images from 50 healthy participants and 300 from 74 patients with CTS were used to train and validate a five‐layer U‐Net model. Each CSA measurement was performed in triplicate. Model performance was evaluated using the Dice similarity coefficient (DSC) and by comparing automated with manually annotated CSA values. Prospective evaluation was conducted on a small, unseen dataset and on data acquired using a different ultrasound system.

Results

Manual CSA measurements showed excellent repeatability (ICC 0.982). The model achieved high DSCs (0.95 in controls, 0.96 in CTS) and showed no statistically significant difference from manual CSA measurements ( p = 0.227). Segmentation inaccuracies in the test set were primarily attributable to minor contouring differences at the epineurial border. On unseen data, errors occurred in more proximally scanned images or scans showing intraneural hyperechogenicity.

Discussion

In line with current literature, a CNN with a 2D U‐Net architecture demonstrates strong potential for MN segmentation on static ultrasound images. However, reduced accuracy on unseen data suggests overfitting. Further validation is required to ensure robustness across anatomical and technical variation before clinical implementation.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Carpal Tunnel SyndromeMedian NerveNeural Networks, ComputerHumansConvolutional Neural NetworksFemaleUltrasonographyMiddle AgedAdultMaleAgedReproducibility of ResultsProspective StudiesImage Processing, Computer-Assisted

Stichwörter der Autor:innen

Dice similarity coefficient (DSC)carpal tunnel syndrome (CTS)convolutional neural network (CNN)median nerve (MN)ultrasonography (US)

Autor:innen und Institutionen

  • Department of Neurology, Canisius Wilhelmina Hospital, Nijmegen, the Netherlands.
  • Radboud University Medical Centre, Donders Institute for Brain, Cognition, and Behavior, Nijmegen, the Netherlands.
  • Department of Neurology, Rijnstate hospital, Arnhem, the Netherlands.
  • Professionals in NAH, Lochem, the Netherlands.
  • Department of Neurosurgery, Radboud University Medical Centre, Nijmegen, the Netherlands.
  • Department of Neurosurgery, Canisius Wilhelmina Hospital, Nijmegen, the Netherlands.

ORCID-Profile: Tom B G Olde Dubbelink, Erik Willemen, Frouke A P Nijhuis, Ronald H M A Bartels, Wim I M Verhagen

Förderung

  • Canisius Wilhelmina Hospital Science Fund

Zitieren

APA

Olde Dubbelink, T. B. G., Willemen, E., Schippers, B., Meulstee, J., Nijhuis, F. A. P., Bartels, R. H. M. A., & Verhagen, W. I. M. (2026). Automated Segmentation of Median Nerve Cross‐Sectional Area in Healthy Controls and Patients With Carpal Tunnel Syndrome Using a Convolutional Neural Network. Muscle & Nerve. https://doi.org/10.1002/mus.70167

Vancouver

Olde Dubbelink TBG, Willemen E, Schippers B, Meulstee J, Nijhuis FAP, Bartels RHMA, et al. Automated Segmentation of Median Nerve Cross‐Sectional Area in Healthy Controls and Patients With Carpal Tunnel Syndrome Using a Convolutional Neural Network. Muscle & Nerve. 2026. doi:10.1002/mus.70167

BibTeX
@article{OldeDubbelink2026,
  author = {Olde Dubbelink, Tom B G and Willemen, Erik and Schippers, Bas and Meulstee, Jan and Nijhuis, Frouke A P and Bartels, Ronald H M A and Verhagen, Wim I M},
  title = {Automated Segmentation of Median Nerve Cross‐Sectional Area in Healthy Controls and Patients With Carpal Tunnel Syndrome Using a Convolutional Neural Network},
  journal = {Muscle & Nerve},
  year = {2026},
  doi = {10.1002/mus.70167}
}
Kostenlos starten →