Implementation of AI technology for MRI in York MS patients
Project write up
The Magnetic Resonance Imaging in MS (MAGNIMS) group recently provided consensus recommendations to measure brain volume in clinical practice (1). Artificial intelligence (AI) is required to measure volume and can also assist in the analysis of white matter lesions. These may help to guide treatment decisions in Multiple Sclerosis (MS).
I felt this was an excellent opportunity to support the implementation of AI for our patients with MS.
The primary objective is to identify the volume of MRI scans done annually between 2019-2021 to justify the use of AI software and estimate costs.
The secondary objective is to identify possible challenges and propose measures to manage them.
All York MS patients between 2019-2021 were included. Electronic records were used to identify MRI scans performed. An MRI of the brain and (any) spine scan done together were considered 2 separate scans.
A visit was made to York University neuroimaging service to experience AI technology. The experience was fed back to the radiologist and some suggestions were drawn up.
Between 2019 and 2021, there were 559 MS patients under York hospital. There were 300 scans in 2019, 228 in 2020 and 248 in 2021.
At the university visit, there were problems encountered in comparing longitudinal scans for one of our sample patients. This was mainly due to different scanners and protocols over a period of several years. Suggestions were made to standardise MRI protocol for MS follow up and using the same scanner where possible. An audit is planned to monitor the progress of AI implementation.
We feel that the volume of scans justify use of AI at York hospital. Plans are being made to introduce AI technology at York. We hope that measures put into place will help deliver a successful implementation of the AI service.
1. Sastre-Garriga J, Pareto D, Battaglini M, et al. MAGNIMS consensus recommendations on the use of brain and spinal cord atrophy measures in clinical practice. Nat Rev Neurol. 2020;16(3):171-182.
More MS Academy Digital health solutions Projects
Encouraging excellence, developing leaders, inspiring change
MS Academy was established five years ago and in that time has accomplished a huge amount. The six different levels of specialist MS training are dedicated to case-based learning and practical application of cutting edge research. Home to national programme Raising the Bar and the fantastic workstream content it is producing, this is an exciting Academy to belong to.