Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

De novo biosynthesis of the polymethoxylated flavones eupatilin and jaceosidin in Saccharomyces cerevisiae.

Metabolic engineering·2026
Same author

Gravity-Dependent Modulation of Downbeat Nystagmus: Insights From Velocity-Storage Dysfunction.

Annals of clinical and translational neurology·2026
Same author

Multimodal feature fusion model for breast mass malignant risk stratification.

Frontiers in oncology·2026
Same author

Dual-Single-Atom Catalyst with Aptamer-Engineered Confined Reactive Milieu for Spatiotemporally Controlled Selective Conversion.

Advanced materials (Deerfield Beach, Fla.)·2026
Same author

Early Differentiation of Prostatic Intraepithelial Neoplasia via Multiparametric MRI Combined with PSA-derived Parameters

Current medical imaging·2026
Same author

Hydrocarbon generation kinetics and history of source rocks in different salinity environments, Dongpu Depression, Bohai Bay Basin, China.

PloS one·2026

Related Experiment Video

Updated: Aug 30, 2025

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

905

aEYE: A deep learning system for video nystagmus detection.

Narayani Wagle1,2, John Morkos3, Jingyan Liu1

  • 1Department of Biomedical Engineering, The John Hopkins University, Baltimore, MD, United States.

Frontiers in Neurology
|August 29, 2022
PubMed
Summary

Deep learning effectively detects nystagmus in telemedicine videos, even with lower resolutions and sampling rates. This technology can aid automated neurological diagnoses via eye movement analysis.

Keywords:
artificial intelligencedeep learningdizzinesseye movementsmachine learningnystagmustelemedicinevertigo

More Related Videos

Video-oculography in Mice
09:43

Video-oculography in Mice

Published on: July 19, 2012

23.9K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.0K

Related Experiment Videos

Last Updated: Aug 30, 2025

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders
05:49

Author Spotlight: Deciphering Electrical Networks Behind Complex Brain Activities and Disorders

Published on: November 1, 2024

905
Video-oculography in Mice
09:43

Video-oculography in Mice

Published on: July 19, 2012

23.9K
Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention
06:37

Author Spotlight: Addressing Technical and Subjective Challenges in Measuring Classroom Attention

Published on: December 15, 2023

4.0K

Area of Science:

  • Ophthalmology
  • Neurology
  • Artificial Intelligence

Background:

  • Nystagmus identification is difficult for non-experts, especially via telemedicine.
  • Deep learning for video-based eye movement detection is understudied.

Purpose of the Study:

  • Develop and validate a deep learning system (aEYE) for nystagmus detection in video-oculography recordings.
  • Assess the model's performance across various video resolutions and sampling rates.

Main Methods:

  • Trained a deep learning model (aEYE) on video-oculography data from the AVERT clinical trial.
  • Tested model performance using video clips with altered sampling rates (60 Hz, 30 Hz, 15 Hz) and resolutions.
  • Utilized a filtered image-based motion classification approach.

Main Results:

  • An ensemble model achieved an AUROC of 0.86, sensitivity of 88.4%, specificity of 74.2%, and accuracy of 82.7%.
  • Validated folds showed an average AUROC of 0.86, sensitivity of 80.3%, specificity of 80.9%, and accuracy of 80.4%.
  • Accuracy decreased with lower sampling rates, but was minimally affected by reduced image resolution.

Conclusions:

  • Deep learning is effective for nystagmus detection in videos, including those with reduced quality.
  • aEYE shows potential as a tool for automated, eye-movement-based neurological diagnosis in telemedicine settings.