Semiautomated breast ultrasound report generation using multimodal large language models and deep learning
The proposed framework enables accurate, efficient, and device-adaptable breast US report generation by combining multimodal DL and prompt-based LLM inference. It significantly reduces radiologist workload and demonstrates potential for scalable deployment in real-world clinical workflows.
Authors
- Sung Eun Song (ORCID: https://orcid.org/0000-0002-9259-8294)
- Kyu Ran Cho (ORCID: https://orcid.org/0000-0002-8936-6468)
- Khadija Azhar
- Byoung-Dai Lee
- Shi Sub Byon
- SeungJae Lee
Institutions
- Korea University (KR)
- Kyonggi University (KR)
- Cohu (United States) (US)
- Institut Régional d'Administration de Bastia (FR)
Publication Details
- Journal
- Frontiers in Medicine
- Published
- 2026-01-21
- DOI
- https://doi.org/10.3389/fmed.2026.1679203
- Primary Topic
- AI in cancer detection
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Korea Health Industry Development Institute