Facebook’s FAIR AI research lab has reportedly inked a collaboration with the NYU School of Medicine’s Department of Radiology to revolutionize magnetic resonance imaging (MRI) scans. Sources claim that the new research project by Facebook aims to make MRI scans up to 10 times faster through the use of AI technology.
The project has plausibly been initiated on the grounds of the fact that MRI scanning, a long-drawn-out process, easily takes up 60 minutes or more of a patients time. With the help of AI, this project aims to cut down the prolonged process by scanning only what’s important. If initiated successfully, the project can improve the speed of scans and help patients access the service for a lesser cost, stated sources familiar with the development.
According to Engadget, the project has access to about 3 million MRI images and 10,000 clinical cases. The team plans to make baselines, AI models, image data sets, and evaluation metrics available to other researchers as well.
Facebook confirmed that AI can help scan faster while capturing lesser data without losing significant information. The notion here is to suitably design artificial neural networks that would recognize the underlying structure of the images to fill in the view omitted from the accelerated scan.
As per sources familiar with the matter, the project is Facebook’s attempt to apply its experimental AI research to real-world problems, including the ones in the medical space. Incidentally, the social media giant was recently suspected of initiating talks with major hospitals and medical groups to share data of their most vulnerable patients. Google’s AI lab had also been accused of violating British privacy laws due to its data-sharing partnership with the UK’s National Health Service.
Sources cite that with this project, Facebook aims to overturn the outrage by affirming that no data from the company’s social networking platform will be used in the MRI initiative.