Brain‐FM: A Multimodal Foundation Model for Visual Question Answering in Brain Health Diagnostics
This study introduces Brain-FM, a multimodal foundation model tailored for brain health applications, leveraging visual-text data to enhance diagnostic reasoning. Experimental results show that mixing the visual features achieves high accuracy when finetuned on PMC-Dementia dataset. Brain-FM represents a significant step toward more flexible, generalizable AI systems for brain health diagnostics and reasoning.
Authors
- Tianze Yu (ORCID: https://orcid.org/0000-0003-3800-3912)
- Xuling Lin (ORCID: https://orcid.org/0000-0003-1281-1145)
- K. -M. Ng
- Somayeh Ebrahimkhani
- Lok Yee Tweety Au
- Simon Kang Seng Ting (ORCID: https://orcid.org/0000-0003-4852-9790)
- Adeline Su Lyn Ng
- Shahul Hameed
- Eng King Tan
- Guimeng Liu
- Ngai‐Man Cheung
Institutions
- Singapore University of Technology and Design (SG)
- National Neuroscience Institute (SG)
Publication Details
- Journal
- Alzheimer s & Dementia
- Published
- 2025-12-01
- DOI
- https://doi.org/10.1002/alz70856_104430
- Primary Topic
- Multimodal Machine Learning Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00