Back

Privacy-Preserving Generative AI Framework for Synthetic Multimodal Eye Movement Data

Rahman, A.; Bachina, P.; Patel, V.; Green, K. E.

2025-04-22 bioengineering
10.1101/2025.04.21.649786 bioRxiv
Show abstract

Eye movements, such as nystagmus, saccades, and smooth pursuit, provide valuable information about neurological function but have limited publicly accessible datasets due to patient privacy concerns. To address this, we leverage generative AI to create realistic videos of artificial eye movement, eliminating the need for real patient data. These synthetic datasets have shown performance comparable to actual patient data in clinical tasks. Our generated videos will be openly shared, facilitating broader research and advancement in neurologic and neuro-ophthalmic AI applications.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.