Together with Philips, Leiden researchers have won a competition in which international research groups are working on accelerating MRI scans with the help of artificial intelligence (AI). The LUMC and Philips team developed an algorithm that can reconstruct an almost as good MRI image of a knee with eight times less data than normal
In the 'fastMRI challenge', issued by the Facebook AI research lab and New York University, AI specialists were given the challenge of making MRI scans faster and more efficient. The 34 participating teams received a raw data set of a few hundred MRI scans of knees, as well as a number of incomplete data sets. Based on this they had to develop a self-learning algorithm, with which the complete MRI image can still be calculated from the incomplete data sets. Recording less data during the MRI scan means a short scan time.
The Philips and LUMC team ended up at the top of the most difficult part of the competition: an almost intact picture taken from just one eighth of the data. According to the LUMC, a considerable group of researchers worked full-time for months on the self-learning algorithm. AI expert Marius Staring and MRI specialist Thijs van Osch (both Radiology department) were the team leaders from LUMC.
Staring mentions the close cooperation between AI experts, MRI technicians and clinicians, from both the LUMC and Philips , as key to the success. “The collaboration with Philips was very intensive, so we were able to quickly test all kinds of ideas. I am proud of the LUMC researchers Sahar Yousefi and Mohamed Elmahdy who, together with colleagues at Philips, have delivered a top performance. ”
AI PROMISE FOR IMAGING WORLD
The researchers emphasize the great promise that AI technology holds for the imaging world. “Currently, an MRI scan can easily take 15 minutes to 30 minutes. If we can reduce that to a few minutes, that will be a benefit for both the patient and the practitioner, “Van Osch says. “We look forward to continuing the collaboration with Philips, both to conduct further research and also to bring this profit to the patient quickly.”AI, Art, gezondheid, gezondheidszorg, kosten besparen, ml, mr, ziekenhuis