Exploring Generative Pre-Trained Transformer-4-Vision for Nystagmus Classification: Development and Validation of a Pupil-Tracking Process
We demonstrated the potential of versatile vertigo management in a generative artificial intelligence model that improves the accuracy and efficiency of nystagmus classification. We also highlighted areas for further improvement, such as expanding the dataset size and enhancing input modalities, to improve classification performance across all nystagmus types. The GPT-4V model validated only for recognizing still images can be linked to video classification and proposed as a novel method.
Authors
- Masao Noda (ORCID: https://orcid.org/0000-0001-6281-2782)
- Hirofumi Ogihara (ORCID: https://orcid.org/0000-0001-8335-2279)
- Tomohiko Kamo (ORCID: https://orcid.org/0000-0003-0490-8393)
- Hiroaki Fushiki (ORCID: https://orcid.org/0009-0003-4120-8641)
- Ryota Koshu (ORCID: https://orcid.org/0009-0007-0127-9137)
- Reiko Tsunoda (ORCID: https://orcid.org/0000-0001-8151-9071)
- Makoto Itō (ORCID: https://orcid.org/0000-0001-6863-4622)
Institutions
- Jichi Medical University (JP)
- Mejiro University (JP)
Publication Details
- Journal
- JMIR Formative Research
- Published
- 2025-06-06
- DOI
- https://doi.org/10.2196/70070
- Primary Topic
- Vestibular and auditory disorders
- Type
- article
- Field-Weighted Citation Impact
- 0.00