Deep learning-based prediction of individualized Real-time FSH doses in GnRH agonist long protocols
This is the first study to apply a cross-temporal and cross-feature deep learning framework for daily, individualized FSH dose prediction across the full duration of COS. The model demonstrated superior performance over conventional approaches and offers a promising tool for standardizing COS management. Although currently limited by its retrospective, single-center design, the model may support future clinical decision-making and improve COS outcomes. Prospective, multicenter validation studies are warranted to confirm its utility and generalizability.
Authors
- Na Kong (ORCID: https://orcid.org/0000-0002-6613-1957)
- Hui Zhang (ORCID: https://orcid.org/0000-0003-3491-6394)
- Jie Mei (ORCID: https://orcid.org/0000-0002-5182-5059)
- Zhilong Wang (ORCID: https://orcid.org/0000-0003-0658-2614)
- Haixiang Sun (ORCID: https://orcid.org/0000-0002-1215-1792)
- Wu-Jun Li (ORCID: https://orcid.org/0000-0003-2499-4006)
- Chenyang Huang (ORCID: https://orcid.org/0000-0003-2408-4483)
- Yu Xia (ORCID: https://orcid.org/0009-0001-7203-1418)
- Yingchun Zhu
- Liyan Duan
Institutions
- Nanjing University of Chinese Medicine (CN)
- Nanjing University of Science and Technology (CN)
- Nanjing Drum Tower Hospital (CN)
- Nanjing Medical University (CN)
- Nanjing University (CN)
Publication Details
- Journal
- Journal of Translational Medicine
- Published
- 2025-05-15
- DOI
- https://doi.org/10.1186/s12967-025-06562-8
- Citations
- 2
- Primary Topic
- Hypothalamic control of reproductive hormones
- Type
- article
- Field-Weighted Citation Impact
- 3.40
Funders
- National Natural Science Foundation of China
- Government of Jiangsu Province
- Nanjing Drum Tower Hospital