Pocket-Sized $200 Fundus Camera Reduces Pupil Dilation

Troy Brown, RN

March 28, 2017

Researchers have developed an inexpensive, pocket-sized, nonmydriatic fundus camera using mostly off-the-shelf electrical components. The camera will not replace the use of dilating eye drops, but will make a method of nonmydriatic retina imaging already in use more portable and accessible.

"Dilating eye drops are often necessary for patients with small pupils, or for viewing the peripheral parts of the retina," researcher Bailey Y. Shen, MD, from the Illinois Eye and Ear Infirmary, University of Illinois at Chicago, told Medscape Medical News.

"What is interesting...about our prototype is that we took advantage of advances in technology (small and cheap camera boards, [and] cheap [light-emitting diodes] capable of emitting dual infrared and white light) to build a working nonmydriatic camera that is very cheap and portable," Dr Shen explained.

Dr Shen and Shizuo Mukai, MD, from the Massachusetts Eye and Ear Infirmary, Harvard Medical School, Boston, Massachusetts, describe the camera in an article published online March 15 in the Journal of Ophthalmology. "We made our paper open-access...and hope that it will be a step in making nonmydriatic fundus imaging accessible for more people," Dr Shen said.

Source: courtesy Bailey Shen and Shizuo Mukai.

The researchers based the camera on the Raspberry Pi 2 computer, a popular, general purpose, single-board computer platform. They connected the board to a small inexpensive infrared camera and dual infrared and white LEDs.

A lens, a small display screen, and several cables complete the camera. The camera functions by first emitting infrared light, which focuses the camera on the retina. At this time, most retina cameras use white light, which causes the iris to contract, constricting the pupil; thus, dilating eye drops are necessary to be able to visualize the retina.

With the new camera, the infrared light focuses the camera on the retina after a few seconds without provoking a response from the iris. Once the camera is focused, it delivers a quick flash of white light and takes the picture.

Cameras that use this same infrared/white light technique have existed for years, but they are often bulky and cost thousands of dollars.

Dr Shen's and Dr Mukai's camera images the retina and its blood supply, as well as the part of the optic nerve that enters the retina. It can discover health issues including diabetes, glaucoma, and elevated pressure around the brain.


Previous research has demonstrated that smartphones can perform fundus photography, the authors write. Smartphones are ubiquitous and connect easily to mobile or wireless networks. However, when their native camera and flash are used, pharmacologic dilation is necessary, and the safety of the camera flash is unknown among different smartphones.

"[A] major possible improvement to our nonmydriatic fundus camera would be to turn the camera into a 'dongle' for a smartphone, connected to a smartphone via a micro-USB or Lightning [Apple] connection," the authors write. "Such a camera dongle would have an infrared-sensitive camera board and dual infrared and white light LED, but would rely on the smartphone to provide the battery, touchscreen viewfinder, and internet connection, increasing the portability and decreasing the cost of the camera."

Dr Mukai told Medscape Medical News, "Making a retinal imaging device that requires no pharmacologic dilation of the pupil that is inexpensive and portable...will make the device more readily available even in the poorest and most remote setting," If the device were an attachment to a smartphone, it would "also allows for immediate telemedicine capability via the mobile phone platforms and the Internet," he added.

"Nonmydriatic fundus photography is especially useful for screening and telemedicine. Diseases such as diabetic retinopathy and glaucoma are already being evaluated this way. The cost of the commercially available cameras has limited the scope of the telemedicine networks, and dramatic lowering of the cost should significantly expand the range," Dr Mukai said. "By making the device as an attachment to a smartphone, we can use the smartphone as the computer that it is and use machine learning and [artificial intelligence] for automated disease detection (differentiating diseased from normal retina), and even diagnosis. Our collaborators at MIT Media Lab are already starting to do this. In addition, computational processing can improve the captured images by 'stitching' multiple images for a wider angle montage and by 'stacking' multiple images to improve image resolution," he explained.

"Additional Refinements," "Formal Testing" Next Steps

The team is trying to find collaborators who can help them more comprehensively test the light safety of their prototype camera, Dr Shen told Medscape Medical News. He said it is difficult to say how long that will take, as they may want to make additional adjustments to the illumination system in the future.

"There will be additional refinements to the device, and formal testing would be done on the version closer to the 'clinical' version," Dr Mukai said. "We are fortunate to have a collaborator who is the world expert on retinal-imaging device safety, and we should be able to do it as we did for our iPhone 4 smartphone fundus photography system that we published. As mentioned in our paper, the energy levels reaching the retina and neighboring tissues are well within the safe range," he explained.

"I think that once we can confidently say the lighting of the camera is safe for human eyes, it would be ready for real-world use. What's reassuring is that there are already many commercially available nonmydriatic cameras that we know emit safe amounts of infrared and white lights, so theoretically, a portable and cheap nonmydriatic camera should be capable of being safe as well," Dr Shen added.

Open Access

The researchers include a list of parts, assembly instructions, and the necessary code to program the camera. "We are making the instructions for this device open source...with the hope that others will build it and make further modifications in an innovative way," they write.

"In addition to the device itself, the open-source [do-it-yourself] concept is new to medical-device development," Dr Mukai said. "Since normal pathway and timetable for device development takes several steps, almost a decade of time (when technology is often outdated), and even millions of dollars, one of our goals was to have a device from inception to use in a relatively short time. We were able to do this recently for developing a system for taking fundus photographs using a smartphone, although that system did not involve making a new device," he explained.

"Our prototype camera offers a proof of concept that a portable nonmydriatic fundus camera capable of capturing quality fundus photographs can be produced simply and inexpensively," the authors write. "That such a prototype camera is possible is due in large part to the current smartphone and mobile technology revolution, with improving quality, shrinking sizes, and falling prices of computer chips, camera boards, camera illumination systems, and touchscreens." they conclude.

The authors have disclosed no relevant financial relationships.

J Ophthal. Published online March 15, 2017. Full text

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.