Sensitive visual field assessment using augmented reality
JHU Ref #: C17601
Unmet Need
While visual field measurement is key in diagnosing and monitoring eye and systemic diseases in patients (American Academy of Ophthalmology, 2022), current practices are either quick but imprecise (confrontation visual field tests that rely on the doctor’s hand gestures and placements), or sensitive but expensive and bulky. The gold standard involves a large machine that employs static perimetry tests, requiring the patient to place their face in a bowl-shaped instrument in-clinic and respond to tracked lights in the perimeter. These devices retail for thousands of dollars and are infeasible for use in emergency situations outside the clinic. Therefore, there is a need for a more affordable and portable mechanism to accurately measure patient visual fields.
Value Proposition:
· Handheld virtual reality device that can be applied to a wide array of injury and disease states that negatively affect visual field
· Fast and easy to deploy
· No need for specialized clinician, allowing use outside the clinic in emergency circumstances
· Low cost versus current similar measures
Technology Description
Researchers at Johns Hopkins University have developed machine learning-based software that is capable of assessing visual field detriments in a fast and portable manner using handheld virtual reality devices. The technology responds to patient input and adapts the remainder of the assessment accordingly, which allows for quick and accurate measurements to enumerate and characterize (location, size, and sensitivity) blind spots and other visual deficits. Its adaptability allows for reports at speeds that far surpass the existing gold standard of visual field measurement, and its application to handheld cell phone or virtual reality headset hardware allows for deployment in settings outside the clinic without a specialized technician, which is key in certain lower-resource environments or in emergency situations.
Stage of Development
Proof-of-concept has been established for this algorithm with sample data.
Data Availability
Data available upon request.
Publication
N/A
