15 Sep 2025
US-developed eye camera captures digitally-refocusable retinal images without mechanical focusing, simplifying eye exams.
Eye exams are an essential tool for detecting diseases such as diabetic retinopathy, glaucoma, and age-related macular degeneration, which together account for much of the world’s vision loss.
A key part of this process is fundus imaging, where doctors photograph the back of the eye to look for early signs of disease. Yet the cameras used for this task are expensive, difficult to operate, and require careful focusing, limiting their availability in many clinics and underserved regions.
A team of researchers from Johns Hopkins University, Boston University, and other collaborators, has now demonstrated a new type of fundus camera that removes one of the main challenges: focusing. Instead of relying on mechanical adjustments of a lens, the system uses a special diffuser that captures 3D light information during imaging.
With this data, a computer can reconstruct and refocus images after they are taken. The work, published in Biophotonics Discovery, presents the first demonstration of this method in living human eyes.
‘A modified commercial fundus camera’
To build the system, the team modified a commercial fundus camera, replacing part of its optics with a holographic diffuser and sensitive digital sensor. They then calibrated the system by recording how the diffuser blurs light with different amounts of refractive error. Once calibrated, the camera could record images of the retina in model eyes and digitally sharpen them to the correct focus using software.
In tests with volunteers, the device successfully captured color images of the retina, showing features such as the optic disc, blood vessels, and macula. Importantly, the images could be refocused across a wide range of refractive errors—more than ten diopters—without moving any optical components.
The system produced consistent resolution of about 7 to 10 line pairs per millimeter, which is lower than that of a conventional camera but remained stable for eyes with refractive errors, all with fixed optics and no prior focusing.
The results show that diffuser-based computational imaging can work in real-world eye exams. Because the approach avoids moving parts, it could help reduce the cost and complexity of fundus cameras. The researchers also see potential for combining this technique with autorefractors, which measure eyeglass prescriptions, into a single device. Such a tool could make comprehensive eye exams simpler to perform and more widely accessible.
• This article was first published on spie.org.