Facebook has partnered with New York University (NYU) on fastMRI, a project with the aim of making magnetic resonance imaging (MRI) technology up to 10 times faster by harnessing artificial intelligence (AI), the social media giant has announced.
If the project achieves the desired results, MRI technology would be available to more people, meaning a greater number of diagnosis, according to Facebook.
Currently, MRI scanners take between 15 minutes to an hour, which can be challenging for young children or those with certain disabilities. The length of time also results in long scheduling backlogs, particularly in rural areas with limited access.
Visit Innovation Enterprise's Big Data & Machine Learning for Pharma Summit Europe 2018 in Madrid on October 10, 2018
The university will provide Facebook with access to about 3 million images of MRI scans of the brain, knee and liver from 10,000 clinical cases. The data gathered from the scans will be used to train the algorithms.
"By using AI, we believe it may be possible to capture less data and therefore image faster, while still preserving or even enhancing the rich information content of MR images," said Dr Dan Sodickson, an imaging specialist at NYU.
"The key is to train artificial neural nets to recognize the underlying structure of the images, and what types of things tend to be clustered together, so that the nets can then fill in views that are emitted from the accelerated scan," he added.