Benchmarking Vision LLMs in fetal ultrasound interpretation: a five-point expert evaluation of standard vs. custom prompts
Tailored prompts markedly enhance Vision LLMs' ability to interpret fetal ultrasound images, supporting their potential as aids in prenatal diagnosis and education. However, limitations in anatomical precision and image quality assessment persist. Future research should focus on refining models with specialized datasets, optimizing architectures, and advancing prompt engineering to maximize clinical utility.
Authors
- Marco Agus (ORCID: https://orcid.org/0000-0003-2752-3525)
- A Hamza (ORCID: https://orcid.org/0000-0003-3630-0593)
- Muhammad Tukur (ORCID: https://orcid.org/0000-0003-1103-9659)
- Rao Khan (ORCID: https://orcid.org/0000-0002-7373-6858)
- Mahmood Alzubaidi (ORCID: https://orcid.org/0000-0002-6082-7873)
- S Al Said
- Fatima Anver
- Wala Elsharif (ORCID: https://orcid.org/0000-0003-1933-3254)
- Abdullatif Magram
Institutions
- Hodeidah University (YE)
- Hamad bin Khalifa University (QA)
- Hamad Medical Corporation (QA)
Publication Details
- Journal
- BioMedical Engineering OnLine
- Published
- 2025-12-09
- DOI
- https://doi.org/10.1186/s12938-025-01486-2
- Primary Topic
- Fetal and Pediatric Neurological Disorders
- Type
- article
- Field-Weighted Citation Impact
- 0.00