Christie Medical Physics and Engineering Newsletter 70 May 2022

Acceleration technology If you work in MR you may already have heard of the advanced acceleration technology (AAT) project across the North West. The bid for and allocation of funding was led by Nicola Scott who is the Diagnostic Transformation Manager for Imaging in the North West Regional Diagnostic team. The CMPE MR physics group have had a significant role in supporting this project since December 2021 to help ensure that all trusts with eligible MR systems receive the most suitable AAT software for their systems. The AAT has the potential to provide a significant increase in imaging capacity to the region (with no detriment to image quality) and help address the backlog of patients waiting for MR scans. Approximately £4 million has been allocated for the North West, with the majority of implementation taking place in the next financial year. We expect to achieve a >10% increase in capacity at roughly a tenth of the equivalent cost of outsourcing for the same gain, thus offering exceptional value for money. The AAT consists of three main technologies – simultaneous multi-slice (SMS), Compressed Sensing (CS) and artificial intelligence deep learning reconstruction algorithms (see Fig. 3). The anticipated benefits of these techniques will only be realised after an initial investment of time to apply them across imaging protocols with support from MR physicists. For our customers who have benefited from this technology, the CMPE MR physics

group will be working with you over the coming months to ensure that the techniques are implemented and new protocols are optimised. It’s important to be aware that a condition of the funding is that each trust will report back to NHSEI in approximately 12 months' time and be able to demonstrate an increase in capacity. Example 1 of the AAT technologies: below is Air Recon DL deep learning reconstruction algorithm on a GE system – the image on the left is a conventional acquisition taking 2:50min, while the image on the right is the Air Recon DL image taking only 1:28min (image courtesy of GE Healthcare).

Example 2: (split column image below) simultaneous multi-slice on a Siemens MR system showing the potential to reduce scan times with no detriment to image quality (image courtesy of Siemens Healthcare).

10

Copyright © 2022 by The Christie NHS Foundation Trust. All rights reserved.