Researchers have developed an innovative diagnostic tool powered by artificial intelligence that uses a smartphone camera and cloud computing to detect nystagmus—rapid, involuntary eye movements often associated with balance and neurological disorders.
A Low-Cost Remote Diagnostic Alternative
Unlike traditional and expensive diagnostic methods like videonystagmography (VNG)—which can cost over $100,000—the new AI-based tool offers a low-cost, user-friendly alternative for remote diagnostics.
The system tracks 468 facial landmarks in real time, analyzes eye movement velocity, and generates ready-to-review reports for physicians. Early testing shows that the tool performs comparably to advanced medical equipment, expanding the possibilities of telehealth solutions.
How the Diagnostic Tool Works
Artificial intelligence is playing an increasingly vital role in modern medicine, particularly in medical image analysis. However, most existing models rely on static data, limiting their real-time diagnostic ability. To overcome this, researchers from Florida Atlantic University (FAU), along with other collaborators, developed a prototype powered by deep learning to detect nystagmus, a symptom linked to vestibular and neurological disorders.
Unlike traditional tools like VNG or ENG—which are costly, cumbersome, and uncomfortable—the new tool offers a convenient, reliable method for detecting abnormal eye and balance conditions.
Patients can record their eye movements using their smartphones and upload the video to a cloud-based platform. The system then analyzes the footage and provides remote diagnostic feedback from specialists—all without the need for clinic visits.
Precision Eye Tracking with AI
The AI system tracks specific facial points in high resolution to analyze eye motion and assess the slow-phase velocity (SPV)—a key metric used to determine the severity, duration, and direction of nystagmus. The results are presented through easy-to-interpret charts and reports for healthcare professionals.
Promising Results and Ongoing Development
In a preliminary study published in Cureus and involving 20 participants, the AI tool's results were closely aligned with those from traditional medical equipment, validating its clinical accuracy.
Dr. Ali Danesh, lead researcher and professor at FAU, stated:
“Our AI model offers a promising alternative that can complement—or in some cases replace—traditional diagnostic methods, especially in telemedicine environments.”
He added:
“By integrating deep learning with cloud computing, we make remote diagnostics more flexible, affordable, and accessible, especially for patients in rural and underserved areas.”
The system was trained on over 15,000 videos capturing a wide range of eye movements, and it uses smart filtering algorithms to exclude normal eye activity like blinking for more accurate readings.
A Tool for Modern Telehealth
Beyond diagnostics, the tool helps streamline clinical workflows. Physicians can access AI-generated reports through telehealth platforms, compare them with electronic health records (EHRs), and design personalized treatment plans.
Patients benefit from fewer in-person visits, lower costs, and easier follow-up assessments by uploading new video clips to monitor their condition over time.
Dr. Harshal Sanghvi, postdoctoral fellow at FAU’s College of Medicine and first author of the study, noted:
“Though our technology is still in its early stages, it has the potential to redefine care for patients with vestibular and neurological disorders. Its non-invasive and real-time analysis makes it highly scalable—from clinics and ERs to patients’ homes.”
The research team is now focused on refining the model’s accuracy, expanding clinical trials across broader demographics, and obtaining FDA approval for widespread medical use.
(window.globalAmlAds = window.globalAmlAds || []).push('admixer_async_509089081')
(window.globalAmlAds = window.globalAmlAds || []).push('admixer_async_552628228')
Read More
Security Flaw Exploits Air-Gapped Computers Using Smartwatches
SteelSeries Launches “Arctis Nova 3 Wireless” Headset with Smart Sound App
Threads Finally Adds Direct Messaging After Long Wait
Researchers have developed an innovative diagnostic tool powered by artificial intelligence that uses a smartphone camera and cloud computing to detect nystagmus—rapid, involuntary eye movements often associated with balance and neurological disorders.
A Low-Cost Remote Diagnostic Alternative
Unlike traditional and expensive diagnostic methods like videonystagmography (VNG)—which can cost over $100,000—the new AI-based tool offers a low-cost, user-friendly alternative for remote diagnostics.
The system tracks 468 facial landmarks in real time, analyzes eye movement velocity, and generates ready-to-review reports for physicians. Early testing shows that the tool performs comparably to advanced medical equipment, expanding the possibilities of telehealth solutions.
How the Diagnostic Tool Works
Artificial intelligence is playing an increasingly vital role in modern medicine, particularly in medical image analysis. However, most existing models rely on static data, limiting their real-time diagnostic ability. To overcome this, researchers from Florida Atlantic University (FAU), along with other collaborators, developed a prototype powered by deep learning to detect nystagmus, a symptom linked to vestibular and neurological disorders.
Unlike traditional tools like VNG or ENG—which are costly, cumbersome, and uncomfortable—the new tool offers a convenient, reliable method for detecting abnormal eye and balance conditions.
Patients can record their eye movements using their smartphones and upload the video to a cloud-based platform. The system then analyzes the footage and provides remote diagnostic feedback from specialists—all without the need for clinic visits.
Precision Eye Tracking with AI
The AI system tracks specific facial points in high resolution to analyze eye motion and assess the slow-phase velocity (SPV)—a key metric used to determine the severity, duration, and direction of nystagmus. The results are presented through easy-to-interpret charts and reports for healthcare professionals.
Promising Results and Ongoing Development
In a preliminary study published in Cureus and involving 20 participants, the AI tool's results were closely aligned with those from traditional medical equipment, validating its clinical accuracy.
Dr. Ali Danesh, lead researcher and professor at FAU, stated:
“Our AI model offers a promising alternative that can complement—or in some cases replace—traditional diagnostic methods, especially in telemedicine environments.”
He added:
“By integrating deep learning with cloud computing, we make remote diagnostics more flexible, affordable, and accessible, especially for patients in rural and underserved areas.”
The system was trained on over 15,000 videos capturing a wide range of eye movements, and it uses smart filtering algorithms to exclude normal eye activity like blinking for more accurate readings.
A Tool for Modern Telehealth
Beyond diagnostics, the tool helps streamline clinical workflows. Physicians can access AI-generated reports through telehealth platforms, compare them with electronic health records (EHRs), and design personalized treatment plans.
Patients benefit from fewer in-person visits, lower costs, and easier follow-up assessments by uploading new video clips to monitor their condition over time.
Dr. Harshal Sanghvi, postdoctoral fellow at FAU’s College of Medicine and first author of the study, noted:
“Though our technology is still in its early stages, it has the potential to redefine care for patients with vestibular and neurological disorders. Its non-invasive and real-time analysis makes it highly scalable—from clinics and ERs to patients’ homes.”
The research team is now focused on refining the model’s accuracy, expanding clinical trials across broader demographics, and obtaining FDA approval for widespread medical use.
