Sep 11, 2017 | By Benedict

Researchers at the University of Washington have developed a 3D printed deep learning device called “PupilScreen” that can determine whether a person requires medical attention for concussion or other traumatic brain injuries (TBIs).

Although diagnosed as a “minor” brain injury, concussions are an extremely serious topic in the world of sports and elsewhere. Even with protective headgear, athletes like American footballers and rugby players are extremely susceptible to concussion, a kind of injury that causes a temporary loss of brain function, headaches, and a general feeling of fogginess.

But those are just the short-term effects. What’s really under scrutiny these days is the long-term effects of concussion, especially the effects of multiple concussions over a period of time. Studies suggest that those who have suffered such injuries numerous times are more likely to develop clinical depression and even memory loss conditions like Alzheimer’s disease.

Preventing concussion is therefore incredibly important, and that’s why the development of high-level protective headgear for athletes—but also soldiers, workers, and other people at risk of brain trauma—is so crucial. (3D scanning has been used by helmet specialist Riddell on some of its recent models, while new 3D printed materials could also improve protective headgear.)

You can’t always prevent concussion, though, and the next best solution to the problem is being able to spot the signs of TBIs as soon as possible. Doctors and physiotherapists have the skills to do this, but most of us don’t.

That’s why PupilScreen, a new smartphone app and 3D printed device developed at the University of Washington, could be so useful.

“The vision we’re shooting for is having someone simply hold the phone up and use the flash,” explains researcher Alex Mariakakis. “We want every parent, coach, caregiver or emergency medical technician who is concerned about a brain injury to be able to use it on the spot without needing extra hardware.”

According to the researchers, PupilScreen uses a smartphone’s video camera and deep learning software to see how a person’s eyes are responding to light—a potential indicator of whether or not that person may have concussion.

Using the smartphone’s camera flash, PupilScreen can stimulate the eye, before recording an eight-second video of the pupil’s response. Neural networks then process the video footage, tracking the pupil diameter over the course of those eight seconds and delivering useful numbers—constriction velocity, magnitude of diameter change, etc.—that can be used for diagnosis.

That’s how the smartphone bit works, but the 3D printed attachment is also incredibly important. The simple box-like structure works to block out ambient light, enabling standardization and maximizing the effect of the camera flash. The 3D printed part also keeps the smartphone at a precise distance from the eye.

The researchers say the trickiest part of developing PupilScreen was training the software to distinguish between pupil and iris. Around 4,000 images of eyes had to be annotated by hand in order to “teach” the system which part was which. Now, however, the machine may be smarter than the people that built it, since a computer can quantify subtle changes in the pupillary light reflex that are imperceivable with our own eyes.

Early testing suggests that the 3D printed smartphone system is doing its job. A pilot study involving six patients with TBIs resulted in PupilScreen correctly differentiating healthy patients from injured ones every time. A bigger study involving coaches, emergency medical technicians, and doctors is planned for this fall.

A commercial version of PupilScreen could be available within two years, but the researchers still have some way to go before the device is ready. At present, the system is best at identifying serious TBIs, but struggles at identifying very mild concussion. However, the University of Washington team expects the upcoming fall trials to yield big improvements.

“Having an objective measure that a coach or parent or anyone on the sidelines of a game could use to screen for concussion would truly be a game-changer,” says Professor Shwetak Patel of the University of Washington. “Right now the best screening protocols we have are still subjective, and a player who really wants to get back on the field can find ways to game the system.”

PupilScreen could therefore be the first device to offer a conclusive diagnosis of concussion in non-medical environments.

“PupilScreen…[gives] us the first capability to measure an objective biomarker of concussion in the field,” says Dr Lynn McGrath of the university’s College of Medicine. “After further testing, we think this device will empower everyone from Little League coaches to NFL doctors to emergency department physicians to rapidly detect and triage head injury.”

The PupilScreen team is currently seeking further partners for its next round of testing, which will begin in October.



Posted in 3D Printing Application



Maybe you also like:


Leave a comment:

Your Name:


Subscribe us to Feeds twitter facebook   

About provides the latest news about 3D printing technology and 3D printers. We are now seven years old and have around 1.5 million unique visitors per month.

News Archive