'Eye' on health: AI advances remote detection of dizziness and balance issues

Innovative AI technology from Florida Atlantic University enables remote diagnosis of vestibular and neurological balance disorders using smartphone videos, offering a cost-effective and accessible alternative to traditional methods.
Artificial intelligence is significantly enhancing medical diagnostics, especially in the assessment of balance and vestibular disorders. Researchers from Florida Atlantic University have developed a pioneering deep learning system that enables remote diagnosis of nystagmus—an involuntary eye movement often linked to neurological or vestibular problems—using simple video recordings from smartphones.
Traditional diagnostic methods like videonystagmography (VNG) and electronystagmography are accurate but often expensive, bulky, and inconvenient for patients. In contrast, FAU’s AI-powered platform offers a cost-effective, non-invasive alternative that facilitates quick screenings. Patients can record their eye movements at home, securely upload the videos to a cloud system, and receive expert analysis without visiting a clinic.
This innovative system employs facial landmark tracking—mapping 468 points on the face—and analyzes eye movement velocity to identify signs of nystagmus. The AI provides clinicians with detailed graphs and reports that support virtual consultation processes. A pilot study with 20 participants demonstrated that the AI assessments closely match those made with conventional devices, underscoring the model's initial reliability and potential for clinical integration.
"Our AI tool can complement or replace standard diagnostic devices, especially in environments where access to specialized care is limited," said lead researcher Dr. Ali Danesh. "By combining deep learning, cloud computing, and telehealth, we aim to make diagnosis more accessible and affordable, particularly in rural or underserved communities."
In addition to the software platform, FAU researchers are exploring wearable headsets equipped with deep learning sensors to detect nystagmus in real-time. Though still in early testing stages, these devices promise continuous, non-invasive monitoring, which could revolutionize patient management for vestibular and neurological disorders.
This interdisciplinary effort involves collaborations across FAU colleges and medical centers, with ongoing enhancements to improve accuracy and patient coverage. The goal is to obtain FDA approval for broader clinical use. As telemedicine advances, such AI-driven tools are poised to streamline diagnostics, improve early detection, and reduce healthcare disparities by enabling remote assessment and follow-up care.
Stay Updated with Mia's Feed
Get the latest health & wellness insights delivered straight to your inbox.
Related Articles
Enhanced Imaging Methods Detect Earliest Breast Cancers in Women with Dense Breasts
Advanced imaging techniques like MRI and contrast mammography significantly improve early breast cancer detection in women with dense breast tissue, especially when traditional mammograms are negative. This research highlights the potential for more effective, personalized screening strategies.
How Sound Influences Brainwaves and Reshapes Neural Networks in Real Time
New research reveals how listening to sounds and music can actively induce real-time reorganization of brain networks, advancing our understanding of neural dynamics and connectivity.
Disparities in Access to Advanced Life Support Revealed by Recent Study
Recent research highlights significant disparities in access to ECMO, an advanced life support, revealing systemic biases based on race, income, and hospital proximity. Learn how these inequities impact critical care delivery.