Researchers at the University of Washington developed an app to detect concussions and other brain injuries using a smartphone.
The app, called PupilScreen, uses a smartphone’s flash and analyses recorded video to determine how pupils respond to light.
Typically head injuries are evaluated using a pupilometer, which are usually found in hospitals. Aside from that, other methods involve repeating a list of words or visually examining a pupil’s response with a flashlight. But those methods aren’t error free which is worrisome as brain injuries can have long term effects.
Using deep learning tools, researchers trained the pp to find the pupil of the eye and track its responses to a flash light over a course of three seconds. The smartphone’s camera records the video while the flash is the light source. PupilScreen then produces a report which shows if the pupil’s response time is within normal ranges or otherwise. A pilot study conducted by the developers on 48 patients resulted in accurate detection of brain injury.
The app is currently able to detect severe injuries but refinement and further development is still in progress to enable the app to analyse more complex forms of injuries. Presently the app works with a plastic box, blocking out ambient light and making sure the smartphone is at an appropriate distance from the eye. The team is currently working on making the app useful without any accessories.The smartphone app may not replace traditional MRI machines but it could be a quick to go device for first aid. Emergency personnel could make use of it in case of sports injuries. PupilScreen is to be presented at Ubicomp 2017 this month.