Benchmarking deep learning pipelines for surgical instrument segmentation in endoscopic spine surgery: cross-dataset evaluation under deployment-realistic conditions
European Spine Journal · 21.07.2026
DOI: 10.1007/s00586-026-10162-5 ↗Kurz gefasst
Fachartikel in der Zeitschrift European Spine Journal, online erschienen am 21.07.2026. Verfasst von Bong-Su Mun und 6 weiteren Autor:innen. Beteiligte Einrichtung: Department of Orthopaedic Surgery, Spine Center, Hallym University Sacred Heart Hospital and Hallym University College of Medicine, Anyang, Republic …. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Seoul National University.
Abstract (Originalsprache)
Purpose
Automated surgical instrument segmentation is a prerequisite for AI-assisted guidance in endoscopic spine surgery. Deployment-realistic comparisons of foundation-model and conventional deep learning architectures — accounting for the detector dependency of bbox-prompted models — remain lacking.
Methods
Seven deep learning pipelines were evaluated for binary instrument segmentation: U-Net, Attention U-Net, U-Net++, SegFormer-B0, DeepLabV3+, nnU-Net, and a fully automatic two-stage foundation-model pipeline (MedSAM2-Tiny segmentor + YOLOv8-nano detector). Strict patient-level data splitting was applied across two independent datasets (SNUBH biportal, n = 56 patients; SEA uniportal, n = 60 patients) with three-tier testing (internal, temporal external, cross-dataset external) and two training variants (biportal-only; combined biportal–uniportal).
Results
With biportal-only training, the fully automatic foundation-model pipeline ranked last on cross-dataset external testing (DSC 0.776; 0.737 under a strict empty-mask convention), about 9% points below DeepLabV3+ (0.865). The same MedSAM2-Tiny segmentor reached DSC 0.946 with ground-truth bounding-box prompts, leaving a 16.9-point oracle-to-deployment gap driven by detector domain mismatch: 61% missed detections and 39% box-localization error. With combined training, all seven pipelines exceeded DSC 0.94; nnU-Net was highest (0.978), and the foundation-model pipeline (0.962) was not significantly different from SegFormer-B0 (0.960) or DeepLabV3+ (0.958) after Bonferroni correction, although the small external sample does not establish equivalence. The improvement was driven primarily by detector rather than segmentor learning, and all seven main pipelines except the nnU-Net 5-fold ensemble met the 30 frames-per-second real-time threshold.
Conclusions
Foundation-model two-stage pipeline performance depends critically on detector training data coverage; once the segmentor approaches its oracle ceiling, detector domain coverage becomes the limiting factor for fully automatic deployment. When combined training is feasible, foundation-model and conventional architectures converge to comparable accuracy, although these rankings are specific to the datasets studied and require validation on independent cohorts. Reporting deployment DSC alongside the ground-truth bounding-box upper bound is recommended for future surgical-segmentation benchmarks.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Orthopaedic Surgery, Spine Center, Hallym University Sacred Heart Hospital and Hallym University College of Medicine, Anyang, Republic of Korea.
- Department of Orthopaedic Surgery, Seoul National University Bundang Hospital and Seoul National University College of Medicine, Seongnam, Republic of Korea. psmini@snu.ac.kr.
- Department of Orthopaedic Surgery, Korea University Ansan Hospital and Korea University College of Medicine, Ansan, Republic of Korea.
- Department of Orthopaedic Surgery, Kangnam Sacred Heart Hospital and Hallym University College of Medicine, Seoul, Republic of Korea.
- Department of Orthopaedic Surgery, Seoul National University Bundang Hospital and Seoul National University College of Medicine, Seongnam, Republic of Korea.
Förderung
- Seoul National University
Zitieren
APA
Mun, B. S., Jang, S. I., Park, S. M., Park, J., Park, H. J., Kim, H. J., & Yeom, J. S. (2026). Benchmarking deep learning pipelines for surgical instrument segmentation in endoscopic spine surgery: cross-dataset evaluation under deployment-realistic conditions. European Spine Journal. https://doi.org/10.1007/s00586-026-10162-5
Vancouver
Mun BS, Jang SI, Park SM, Park J, Park HJ, Kim HJ, et al. Benchmarking deep learning pipelines for surgical instrument segmentation in endoscopic spine surgery: cross-dataset evaluation under deployment-realistic conditions. European Spine Journal. 2026. doi:10.1007/s00586-026-10162-5
BibTeX
@article{Mun2026,
author = {Mun, Bong-Su and Jang, Seok-In and Park, Sang-Min and Park, Jiwon and Park, Hyun-Jin and Kim, Ho-Joong and Yeom, Jin S},
title = {Benchmarking deep learning pipelines for surgical instrument segmentation in endoscopic spine surgery: cross-dataset evaluation under deployment-realistic conditions},
journal = {European Spine Journal},
year = {2026},
doi = {10.1007/s00586-026-10162-5}
}
EUVIMED