Aug. 21 (UPI) -- Facebook and the New York University School of Medicine launched a research project using artificial intelligence to improve the speed of MRI scans.
Facebook announced the collaborative effort with the NYU school's Department of Radiology, stating they seek to make MRI, or magnetic resonance imaging, scans up to 10 times faster in order to make them available to more people.
In the post, Facebook said MRI scans provide more detailed information than other medical scans, but can take from 15 minutes up to one hour as opposed to X-Rays and CT scans, which take up to a minute. The lengthy scan time can make MRIs difficult for patients who have difficulty lying down in a confined space.
"By boosting the speed of MRI scanners, we can make these devices accessible to a greater number of patients," Facebook said. "Sufficiently accelerated MRI devices could also reduce the amount of time patients must hold their breath during imaging of the heart, liver, or other organs in the abdomen and torso."
The research aims to use AI to capture less data in order to cut down the total scan time.
"The key is to train artificial neural networks to recognize the underlying structure of the images in order to fill in views omitted from the accelerated scan," Facebook said. "This approach is similar to how humans process sensory information. When we experience the world, our brains often receive an incomplete picture -- as in the case of obscured or dimly lit objects -- that we need to turn into actionable information."
Facebook noted "a few missing or incorrectly modeled pixels" could cause the AI scan to miss a possible ailment, but the streamlined process "can quite literally save lives" by making scans more accessible and cutting costs.
NYU collected 10,000 clinical cases comprising approximately 3 million magnetic resonance images of the knee, brain, and liver to construct the imaging data for the project.
In the wake of the Cambridge Analytica data breach, Facebook confirmed the work is fully HIPAA-compliant, data is fully stripped of patient names and all other protected health information and no Facebook data of any kind will be used.