Brightlamp, a startup out of Purdue University, is developing an app that uses machine learning and the smartphone camera to help diagnose a concussion in about five seconds. Concussions are a type of brain injury that can happen during a collision or impact, causing the affected person to feel dizzy or disoriented. In sports like American Football or boxing, concussions are a common type of injury. Unfortunately, concussions can increase the likelihood of depression and neurodegenerative disease in later life.
If someone sustains an impact on the playing field, it is important to find out if they have suffered a concussion, as continuing to play could cause further injury. One way to diagnose a concussion is to measure how the iris in the eye responds to light.
The Brightlamp app uses the flash on a camera phone to shine light in the eye and then the camera measures pupil dilation. After being sent to the cloud, the data are analysed using machine learning techniques to see if the person shows any of the characteristics of a concussion.
Brightlamp claims that the test is the fastest available and has a 98% accuracy. The idea is that the public can test themselves easily and rapidly if they suspect a concussion, and then look for medical help if required. The company is working on getting FDA approval for the app, which is not yet available commercially.