Large‐scale convolutional neural network for clinical target and multi‐organ segmentation in gynecologic brachytherapy via multi‐stage learning
The proposed multi-stage learning strategy effectively enhances segmentation accuracy for gynecologic brachytherapy, leveraging large-scale self-supervised pretraining and progressive finetuning. By improving HRCTV and OARs delineation, GynBTNet has the potential to enhance treatment planning precision, minimize radiation exposure to critical structures, and improve patient outcomes.
Authors
- Keyur D Shah (ORCID: https://orcid.org/0000-0002-0886-655X)
- Priyanka Kapoor (ORCID: https://orcid.org/0000-0002-7676-9743)
- Richard L. J. Qiu (ORCID: https://orcid.org/0000-0002-7877-1900)
- Mingzhe Hu
- Yuan Shao (ORCID: https://orcid.org/0000-0001-6697-4885)
- Jill Remick (ORCID: https://orcid.org/0000-0002-0241-7622)
- Chih‐Wei Chang (ORCID: https://orcid.org/0000-0002-3818-4381)
- Zhen Tian (ORCID: https://orcid.org/0000-0002-7026-7582)
- Yuheng Li (ORCID: https://orcid.org/0009-0001-0249-8952)
- Yuan Gao (ORCID: https://orcid.org/0000-0002-6988-4217)
- Justin Roper
- Beth Bradshaw
- Xiaofeng Yang (ORCID: https://orcid.org/0000-0001-9920-4641)
Institutions
- Illinois Department of Public Health (US)
- Emory University (US)
- University of Illinois Urbana-Champaign (US)
- Chicago Department of Public Health (US)
- University of Illinois Chicago (US)
- University of Chicago (US)
Publication Details
- Journal
- Medical Physics
- Published
- 2025-08-01
- DOI
- https://doi.org/10.1002/mp.18067
- Primary Topic
- Medical Imaging and Analysis
- Type
- article
- Field-Weighted Citation Impact
- 0.00