MRAnnotator: multi-anatomy and many-sequence MRI segmentation of 44 structures
Radiology Advances · 14.01.2025 · 2 (1)
DOI: 10.1093/radadv/umae035 ↗Kurz gefasst
Fachartikel in der Zeitschrift Radiology Advances, online erschienen am 17.12.2024. Verfasst von Alexander Zhou und 18 weiteren Autor:innen. Beteiligte Einrichtung: BioMedical Engineering and Imaging Institute, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, United States. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Purpose
To develop a deep learning model for multi-anatomy segmentation of diverse anatomic structures on MRI. Materials and Methods In this retrospective study, 44 structures were annotated using a model-assisted workflow with manual human finalization in 2 curated datasets: an internal dataset of 1518 MRI sequences (843 patients) from various clinical sites within a health system, and an external dataset of 397 MRI sequences (263 patients) from an independent imaging center for benchmarking. The internal dataset was used to train an nnU-Net model (MRAnnotator), while the external dataset evaluated MRAnnotator’s generalizability across significant image acquisition distribution shifts. MRAnnotator was further benchmarked against an nnU-Net model trained on the AMOS dataset and 2 current multi-anatomy MRI segmentation models, TotalSegmentator MRI (TSM) and MRSegmentator (MRS). Performance throughout was quantified using the Dice score.
Results
MRAnnotator achieved an overall average Dice score of 0.878 (95% CI: 0.873, 0.884) on the internal dataset test set and 0.875 (95% CI: 0.869, 0.880) on the external dataset benchmark, demonstrating strong generalization (P = .899). On the AMOS test set, MRAnnotator achieved comparable performance for relevant classes (0.889 [0.866, 0.909]) to an AMOS-trained nnU-Net (0.895 [0.871, 0.915]) (P = .361) and outperformed TSM (0.822 [0.800, 0.842], P < .001) and MRS (0.867 [0.844, 0.887], P < .001). TSM and MRS were also evaluated on the relevant classes from the internal and external datasets and were unable to achieve comparable performance to MRAnnotator.
Conclusion
MRAnnotator achieves robust and generalizable MRI segmentation across 44 anatomic structures. Future direction will incorporate additional anatomic structures into the datasets and model. Model weights are publicly available on GitHub. The external test set with annotations is available upon request.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- BioMedical Engineering and Imaging Institute, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, United States.
- Department of Diagnostic, Molecular, and Interventional Radiology, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, United States.
- Rotterdam School of Management, Erasmus University Rotterdam, Rotterdam, 3062PA, Netherlands.
- East River Medical Imaging, New York, NY, 10021, United States.
- Department of Radiology, Weill Cornell Medicine, New York, NY, 10065, United States.
- Windreich Department of Artificial Intelligence and Human Health, Icahn School of Medicine at Mount Sinai, New York, NY, 10029, United States.
ORCID-Profile: Alexander Zhou, Zelong Liu, Anthony Yang, Zahi A Fayad, Timothy Deyer, Xueyan Mei
Zitieren
APA
Zhou, A., Liu, Z., Tieu, A., Patel, N., Sun, S., Yang, A., Choi, P., Lee, H. C., Tordjman, M., Deyer, L., Mei, Y., Fauveau, V., Soultanidis, G., Taouli, B., Huang, M., Doshi, A., Fayad, Z. A., Deyer, T., & Mei, X. (2025). MRAnnotator: multi-anatomy and many-sequence MRI segmentation of 44 structures. Radiology Advances, 2(1). https://doi.org/10.1093/radadv/umae035
Vancouver
Zhou A, Liu Z, Tieu A, Patel N, Sun S, Yang A, et al. MRAnnotator: multi-anatomy and many-sequence MRI segmentation of 44 structures. Radiology Advances. 2025;2(1). doi:10.1093/radadv/umae035
BibTeX
@article{Zhou2025,
author = {Zhou, Alexander and Liu, Zelong and Tieu, Andrew and Patel, Nikhil and Sun, Sean and Yang, Anthony and Choi, Peter and Lee, Hao-Chih and Tordjman, Mickael and Deyer, Louisa and Mei, Yunhao and Fauveau, Valentin and Soultanidis, Georgios and Taouli, Bachir and Huang, Mingqian and Doshi, Amish and Fayad, Zahi A and Deyer, Timothy and Mei, Xueyan},
title = {MRAnnotator: multi-anatomy and many-sequence MRI segmentation of 44 structures},
journal = {Radiology Advances},
year = {2025},
volume = {2},
number = {1},
doi = {10.1093/radadv/umae035}
}
EUVIMED