What question did this study set out to answer?

The aim is to assess the effectiveness of video foundation models in capturing 3D retinal structure for improved diagnosis.

March 7, 2026npj Digital MedicineOpen Access

Shifting the retinal foundation models paradigm from slices to volumes for optical coherence tomography

View Full Paper

Ask AI

Bookmark

The aim is to assess the effectiveness of video foundation models in capturing 3D retinal structure for improved diagnosis.
Investigated V-JEPA video foundation model against traditional retinal foundation models.
Fine-tuned models to detect age-related macular degeneration and glaucomatous optic neuropathy.
Utilized five OCT datasets for benchmarking.
Measured performance using Area Under Receiver Operating Characteristic curve (AUROC).
V-JEPA consistently equaled or outperformed traditional image-based models.
Achieved an average AUROC of 0.94, compared to 0.90 for the best image model.
Results were statistically significant with p < 0.001.

No takes yet. Share an insight, caveat, or question.

Research demonstrates improved diagnostic performance using 3D retinal models in ophthalmology, suggesting better outcomes for eye conditions.

Judkiewicz et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Research demonstrates improved diagnostic performance using 3D retinal models in ophthalmology, suggesting better outcomes for eye conditions.