AIRS Medical Showcases Award-Winning MRI Enhancement AI Solution at RSNA 2022

AIRS Medical, Inc. (Seoul, South Korea) is participating in the 108th Scientific Assembly and Annual Meeting of the Radiological Society of North America (RSNA 2022), the world's leading annual imaging forum, at McCormick Place in Chicago held from November 27 to December 1, 2022.

At its RSNA booth, AIRS is showcasing SwiftMR, an FDA 510(k)-cleared AI Imaging solution that enhances MR images acquired under various conditions, contributing to better image quality and patient experience. SwiftMR utilizes conventional MR imaging techniques such as parallel Imaging and compressed sensing, combined with its award-winning deep learning technology. SwiftMR enhances SNR and sharpness of the images, allowing radiologists to read with confidence and ease. SwiftMR was cleared by both the US FDA and the Korea Ministry of Food and Drug Safety (MFDS) in 2021. Since its official commercial launch in Korea in the fourth quarter of 2021, SwiftMR has been installed in more than 100 hospitals, for an average of 30,000 monthly MRI exams and a grand total of more than 280,000 MRI exams.

Along with booth demonstrations, a special presentation is being held by Roh-Eul Yoo, MD, Ph.D., from the Seoul National University Hospital. Dr. Yoo will be sharing her experience working with SwiftMR in various clinical scenarios, along with the results of recent research collaborations proving the power of SwiftMR in providing superior image quality. Due to its physical principles, MR exams inherently require long scan times, which not only present a negative healthcare experience to the patients, but also limit the productivity of healthcare providers. Image acquisition times could be reduced by various methods, but this mostly results in reduced diagnostic image quality. This balance between scan time and image quality has been the center of MR technology innovation for decades.

With the advent of deep learning, fast MRI also faces a new frontier, as obtaining better image quality with significantly less scan time has been made possible. However, developing a viable model for clinical deployment is hindered by a variety of reasons in different dimensions including wide range of contrast mechanisms, difficulties in defining the label and obtaining large-scale high-quality image data to enable the model training process. In this short talk, real-world experiences of utilizing SwiftMR in routine clinical practice and collaborative research will be shared.

Related Links:
AIRS Medical, Inc.