Seoul National University's College of Engineering (Dean Byoungho Lee) announced on November 24 (Wednesday) that a research team led by Professor Jong-Ho Lee of the Department of Electrical and Computer Engineering succeeded in developing a technology that allows artificial intelligence to design electromagnetic waves used in MRI on its own.
 
Unlike X-ray or CT, which uses radiation, MRI uses magnetic fields and electromagnetic waves to image the inside of the human body in three dimensions. Therefore, it is safe from the problem of radiation exposure, but the energy from electromagnetic waves is transmitted to the human body, which means the scanning time may take longer.
 
To solve this problem, Professor Jong-Ho Lee's research team focused on an artificial intelligence technology called deep reinforcement learning that has also been applied to AlphaGo.
 
Deep reinforcement learning is a method of self-learning by exploring the given environment for the targeted behavior of artificial intelligence. For example, the strategy of repeated playing Go in order to win the game can be said to be a method of self-learning.

 
Dong Myung Shin, a Ph.D. student who led this study, built a virtual MRI environment  and designed the electromagnetic waveform created by artificial intelligence to minimize the energy delivered to the human body while maintaining the quality of the image.
 
As a result, the artificial intelligence succeeded in creating a new efficient electromagnetic waveform. In particular, this waveform shows superior characteristics compared to the existing theory-based results developed by experts, suggesting that there is a more effective waveform design method than the design methods developed by experts so far.
 
Meanwhile, the research results were published in Nature Machine Intelligence, a sister journal of the world-renowned academic journal Nature, and Dong Myung Shin will continue his research as a Chief-AI-Officer at RadiSen, an artificial intelligence medical imaging startup based on the research results.