In the 2017 season, NFL players suffered 291 concussions —the most since the league started sharing data in 2012. To curb the high rate of concussions, football rule makers at the professional and college levels have tried everything from penalizing players for hitting with their helmets to adding impartial sideline doctors who can pull players out of the game after a blow to the head.

The unfortunate reality is that none these changes seem to be successfully reducing the number of concussions. Part of the problem is that it can be difficult to even determine whether a player has been concussed or not, but perhaps new diagnostic technology can help.

As the 2018 season gets into full swing, some college teams are keeping a new gadget on the sidelines: a pair of virtual reality goggles designed to diagnose concussions. The Pac-12 conference will actually use the VR goggles in every sport. (Wrestling, not football, has the highest concussion rate , followed by hockey, says a paper in the American Journal of Sports Medicine.)

The Eye-Sync goggles , made by Palo Alto, California-based company SyncThink, work by displaying a dot traveling in a rough circle and tracking the user’s eyes as they follow the movement of the dot. While the goggles track eye motion, the device is really measuring the brain’s ability to predict the dot’s movement, says SyncThink founder Jamshid Ghajar.

Ghajar, a Stanford neurosurgeon and president of the Brain Trauma Foundation, says he was studying how attention relates to brain function when he realized it’s all about timing. “Your brain is always in the past. Anything you sense has already happened,” he says. “By the time you see a tennis ball, it’s already passed. To interact, you have to predict or anticipate, so you’ve learned to predict the movement of the ball so you can hit it.”

Not surprisingly, a concussion affects the brain’s ability to process information from the eyes and predict movement. If an athlete is knocked in the head or a soldier is too close to an IED explosion, and they perform poorly on the goggles’ eye-tracking test, then it’s possible they have suffered a concussion—but not guaranteed.

“Devices don’t diagnose, doctors do,” Ghajar says. Even so, if someone struggles with the goggles, it might be reason enough to pull them off the field. Even a tiny impairment in the ability to predict the movement of a defensive end or a mortal shell could prove costly.

“What we need is a toolkit—eye tracking could be one of the tools,” Ghajar says.

A variety of other diagnostic techniques are also being developed to fill that toolkit. In the spring, researchers announced that certain biomarkers in plasma could accurately predict whether college athletes had sustained a concussion. Other research suggests that measuring changes in the speed of blood flow to the brain might identify concussions. A spit test that measures genetic material in saliva has shown promise identifying concussions in young patients suffering from long-term symptoms, and another study with kids found success using a hearing test .

One piece of technology with potential is a headset that uses computer models and AI to analyze changes in the brain's electrical currents. Developed for use on the sidelines and the front lines by BrainScope, a company based in Bethesda, Maryland, the “brain injury assessment device” is a headband with dangling electrodes that connects to a handheld device. In a couple minutes, it performs an electroencephalogram (EEG) scan to check for bleeding in the brain. The headset lets physicians know if a more robust CT scan is required, says CEO Michael Singer, and it can also track cognitive performance through a test and express results as a percentile.

Like the goggles, the headset is not a definitive test for concussions. “But if you can begin by answering that first question of, is there a bleed in the brain? And then, is there a functional problem? You can then decide on a return to field,” Singer says.

Much of this new research receives funding from the Department of Defense, the NFL or the NCAA—organizations where people are particularly likely to suffer concussions, particularly hindered by a concussion’s impact on their job performance, and particularly reluctant to admit symptoms of a concussion. The reluctance to disclose symptoms can be especially problematic, considering traditional concussion diagnosis relies largely on subjective information provided by the patient, and neither soldiers nor football players want to come off the field.

“Subjective evaluation [of concussions] is actually the most useful, but not if people lie to you,” says Kim Harmon, a sports medicine professor at the University of Washington and chair of the Pac-12 Student Athlete Health and Well-Being Initiative’s board.

Going back out onto the field after suffering a concussion can result in slower reaction times, impaired balance and increased risk of long-term health problems like the degenerative brain disease chronic traumatic encephalopathy (CTE). Faster, more accurate, objective concussion diagnoses present a solution to mitigate the reliance on patient responses.

BrainScope’s Singer compares assessing a concussion to assessing a heart attack. In both cases, a physician must examine a number of tests, check the levels of various compounds and consider the patient’s personal history before reaching a conclusion. In the case of a heart attack, however, doctors needn’t rely much on self-reported symptoms.

“What we have is the need for objective information,” Singer says. “That is the Holy Grail—to have something that cannot be gamed.”

Doctors are often frustrated by the expectation that concussion diagnosis should come quickly. “It’s not like we can wave a wand over them to tell,” Ghajar says. “There’s no green light/red light.”

Unlike other injuries, such as ACL tears or broken wrists, it can be difficult to ascertain whether a person hit in the head is injured at all. Determining whether someone just had their “bell rung,” or if their brain slammed into the side of their skull with enough force to temporarily inhibit cognition, is a tricky science.

“Some people have obvious signs of concussions,” Harmon says. “They can’t stand up straight or tell what’s going on.” But some people are harder to diagnose, retaining most of their faculties after suffering a blow to the head.

Even experts argue about what exactly constitutes a concussion. Ghajar has received funding from the DoD to help come up with a better definition for “concussion,” but in the meantime, researchers tend to use the term “brain injury” to avoid confusion.

“These things are so heterogeneous,” says Sandy Wright, a concussion researcher at the University of British Columbia who has studied how changes in blood flow to the brain could indicate a concussion. Each person can be affected differently by a concussion, he says, and each concussion can harm different parts of the brain or produce different symptoms.

The variability of concussions also affects recovery times, making it difficult to know when a concussed player or soldier is ready to return to action. “I don’t think we’re ever going to come up with one single test that will be the be-all-end-all on either the diagnostic or prognostic side of things,” Wright says.

For Harmon and her colleagues, the exciting part of all this research will come after football season ends. Once players hang up their helmets, researchers will gather all the data from the new tests, including the VR goggles, and information about all the confirmed concussions from the season—which mostly happen in practice—to run a regression analysis. That analysis should let medical scientists know which tools work to actively identify concussions in real time.

There may never be a magic wand to wave, but with new technologies like the Eye-Sync goggles and BrainScope headset, the diagnostic toolkit is growing.