Many patients feel panicky and claustrophobic while getting MRI scans, which involve lying motionless in a large tube for at least 30 minutes while magnetic and radio waves create detailed images of the inside of the body.
Puyang Wang, a doctoral degree candidate in electrical and computer engineering, has a possible solution: an algorithm that speeds up MRI data acquisition, resulting in clearer images in less time. This promising approach was one of several recognized at the fastMRI competition held late last year by Facebook AI and NYU Langone Health.
The competition challenged participants to use artificial intelligence to improve the MRI process. The hope was that by using AI to create accurate images from significantly less raw data, scans could be done as much as 10 times faster, improving the patient experience and making the procedure less expensive.
Wang used recurrent neural networks, a deep learning method that creates clearer images from partial scans. Traditionally, partial scans produce subpar images that are difficult to analyze. However, using Wang’s algorithm, radiologists can get clear results.
“What I like is that it is a very direct way to help the medical industry,” says Wang. “It’s exciting to do this work and see that the real-world application of it could help a lot of people.”