EUVIMEDEUVIMED entdecken
Multizentrische StudieOpen Access · CC BY 4.0

Deep learning-based pelvimetry in pelvic MRI volumes for pre-operative difficulty assessment of total mesorectal excision

Simon C Baltus iD, Ritch T J Geitenbeek, Maike Frieben, Elina Thibeau-Sutre, Jelmer M Wolterink, Can O Tan, Matthijs C Vermeulen, Esther C J Consten, Ivo A M J Broeders

Surgical Endoscopy · 2025

DOI: 10.1007/s00464-024-11485-4 ↗

Kurz gefasst

Multizentrische Studie in der Zeitschrift Surgical Endoscopy, 2025. Verfasst von Simon C Baltus und 8 weiteren Autor:innen. Beteiligte Einrichtung: Surgery Department, Meander Medical Centre, Maatweg, Amersfoort, 3818 TZ, Utrecht, The Netherlands. sc.baltus@meandermc.nl. Hauptthemen laut MEDLINE: Deep Learning, Magnetic Resonance Imaging, Pelvimetry, Preoperative Care. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ethicon.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Background

Specific pelvic bone dimensions have been identified as predictors of total mesorectal excision (TME) difficulty and outcomes. However, manual measurement of these dimensions (pelvimetry) is labor intensive and thus, anatomic criteria are not included in the pre-operative difficulty assessment. In this work, we propose an automated workflow for pelvimetry based on pre-operative magnetic resonance imaging (MRI) volumes.

Methods

We implement a deep learning-based framework to measure the predictive pelvic dimensions automatically. A 3D U-Net takes a sagittal T2-weighted MRI volume as input and determines five anatomic landmark locations: promontorium, S3-vertebrae, coccyx, dorsal, and cranial part of the os pubis. The landmarks are used to quantify the lengths of the pelvic inlet, outlet, depth, and the angle of the sacrum. For the development of the network, we used MRI volumes from 1707 patients acquired in eight TME centers. The automated landmark localization and pelvic dimensions measurements are assessed by comparison with manual annotation.

Results

A center-stratified fivefold cross-validation showed a mean landmark localization error of 5.6 mm. The inter-observer variation for manual annotation was 3.7 ± 8.4 mm. The automated dimension measurements had a Spearman correlation coefficient ranging between 0.7 and 0.87.

Conclusion

To our knowledge, this is the first study to automate pelvimetry in MRI volumes using deep learning. Our framework can measure the pelvic dimensions with high accuracy, enabling the extraction of metrics that facilitate a pre-operative difficulty assessment of the TME.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningMagnetic Resonance ImagingPelvimetryPreoperative CareRectal NeoplasmsPelvic BonesHumansFemaleAnatomic LandmarksMaleMiddle AgedAgedRectumImaging, Three-Dimensional

Stichwörter der Autor:innen

Magnetic resonance imagingPelvimetryTotal mesorectal excisionU-Net

Autor:innen und Institutionen

  • Surgery Department, Meander Medical Centre, Maatweg, Amersfoort, 3818 TZ, Utrecht, The Netherlands. sc.baltus@meandermc.nl.
  • Robotics and Mechatronics, University of Twente, Drienerlolaan, Enschede, 5722 NB, Overijssel, The Netherlands. sc.baltus@meandermc.nl.
  • Surgery Department, Meander Medical Centre, Maatweg, Amersfoort, 3818 TZ, Utrecht, The Netherlands.
  • Surgery Department, University Medical Center Groningen, Hanzeplein, Groningen, 9713 GZ, Groningen, The Netherlands.
  • Department of Applied Mathematics, Technical Medicine Center, University of Twente, Drienerlolaan, Enschede, 5722 NB, Overijssel, The Netherlands.
  • Robotics and Mechatronics, University of Twente, Drienerlolaan, Enschede, 5722 NB, Overijssel, The Netherlands.

ORCID-Profile: Simon C Baltus

Förderung

  • Ethicon

Zitieren

APA

Baltus, S. C., Geitenbeek, R. T. J., Frieben, M., Thibeau-Sutre, E., Wolterink, J. M., Tan, C. O., Vermeulen, M. C., Consten, E. C. J., & Broeders, I. A. M. J. (2025). Deep learning-based pelvimetry in pelvic MRI volumes for pre-operative difficulty assessment of total mesorectal excision. Surgical Endoscopy. https://doi.org/10.1007/s00464-024-11485-4

Vancouver

Baltus SC, Geitenbeek RTJ, Frieben M, Thibeau-Sutre E, Wolterink JM, Tan CO, et al. Deep learning-based pelvimetry in pelvic MRI volumes for pre-operative difficulty assessment of total mesorectal excision. Surgical Endoscopy. 2025. doi:10.1007/s00464-024-11485-4

BibTeX
@article{Baltus2025,
  author = {Baltus, Simon C and Geitenbeek, Ritch T J and Frieben, Maike and Thibeau-Sutre, Elina and Wolterink, Jelmer M and Tan, Can O and Vermeulen, Matthijs C and Consten, Esther C J and Broeders, Ivo A M J},
  title = {Deep learning-based pelvimetry in pelvic MRI volumes for pre-operative difficulty assessment of total mesorectal excision},
  journal = {Surgical Endoscopy},
  year = {2025},
  doi = {10.1007/s00464-024-11485-4}
}
Kostenlos starten →