In-vivo T1, T2 and proton density maps reconstructed with MR-STAT using eight (first column), four (second column), two (third column) and one (fourth column) of the acquired k-spaces in the reconstruction. The acquisition times were 13.6, 6.8, 3.4 and 1.7 seconds respectively on a 1.5T Philips Ingenia MRI system. Find more images and information in this article.
August 10, 2020 — During the International Society for Magnetic Resonance in Medicine (ISMRM) 2020 meeting this week, Philips and University Medical Center (UMC) Utrecht, Netherlands, kicked off a research partnership to advance precision diagnosis through breakthrough quantitative MRI technology. The exclusive, multi-year research partnership will establish a global clinical research network with the aim of fully commercializing the technology, which has initially been developed at UMC Utrecht.
MR-STAT is a major shift in MRI, relying on a new, smart acquisition scheme and machine-assisted reconstruction. MR-STAT delivers multiple quantitative MR parameters in a single fast scan, and represents a significant advance in MR tissue classification, fueling big data algorithms and AI-enabled integrated diagnostic solutions.
In today’s MR systems, quantitative parameters are only produced after the raw data have already been transposed into images. With MR-STAT, the quantitative parameters are reconstructed directly from the raw data. This approach, fueled by recent advances in computing and algorithm development, effectively maximizes the use of every data point collected and removes the need for prior knowledge such as signal dictionaries. Philips said this opens a wealth of opportunities in standardizing quantitative MRI, at very short scan times that are very attractive for clinical practice.
“Based on the feedback from clinical research studies, we strongly believe that this technology can provide tremendous value in the clinic,” said Professor Nico van den Berg, chair of the computational imaging group at UMC Utrecht and one of the technique’s inventors. “We are very pleased with this partnership and look forward to accelerating this technology with more clinical partners in the near future.”
“At Philips, we’re focused on supporting healthcare providers to realize first-time-right diagnosis through clinically relevant and intelligent diagnostics,” said Joland Rutgers, Research and Development Leader for MR at Philips. “With this fast quantitative and single acquisition technology, enhanced with AI, MR-STAT will play a pivotal role in delivering the best diagnostic outcome at an affordable cost, benefitting both healthcare providers and their patients.”
1. Oscar van der Heide, Alessandro Sbrizzi, Peter R. Luijten, and Cornelis A. T. van den Berg. High resolution in-vivo MR-STAT using a matrix-free and parallelized reconstruction algorithm
Submitted to Magnetic Resonance in Medicine on 2019-04-29. Published online at GroundAI.com. https://www.groundai.com/project/high-resolution-in-vivo-mr-stat-using-a-matrix-free-and-parallelized-reconstruction-algorithm/1. Accessed Aug 10, 2020.