Deep learning pipeline for trapezium segmentation in thumb radiographs
A two-stage AI pipeline (YOLOv8 + U-Net) accurately detects and segments the trapezium on thumb radiographs. The method outperforms popular segmentation models and achieves expert-level reproducibility. This tool may enhance surgical planning and intraoperative guidance for TMC arthroplasty.
Authors
- Mélanie Courtine (ORCID: https://orcid.org/0000-0002-9892-613X)
- Thomas Grégory (ORCID: https://orcid.org/0000-0001-6292-0177)
- Younès Bennani (ORCID: https://orcid.org/0000-0003-3667-3357)
- Youssef FRIKEL
- Victor Maigné
- Félix Barbier
Institutions
- Université Sorbonne Paris Nord (FR)
- Assistance Publique – Hôpitaux de Paris (FR)
- Laboratoire des Sciences des Procédés et des Matériaux (FR)
- Hôpital Avicenne (FR)
Publication Details
- Journal
- European Radiology Experimental
- Published
- 2026-02-23
- DOI
- https://doi.org/10.1186/s41747-026-00678-2
- Primary Topic
- Orthopedic Surgery and Rehabilitation
- Type
- article
- Field-Weighted Citation Impact
- 0.00