By harnessing large amounts of complex medical data more efficiently, artificial intelligence (AI) could help discover new disease mechanisms, support early diagnosis, and tailor treatment strategies for chronic conditions such as multiple sclerosis (MS).
In this project, Associate Professor Chenyu Wang and Professor Michael Barnett will use the power of AI to analyse magnetic resonance imaging (MRI) scans. This would enable early intervention to prevent future disability. These tools could transform traditional MRI reports into accurate measures of disease progression.
The team’s second aim is to design a system that can use AI to interpret MRI analyses in appropriate clinical settings, by comparing scans against people with the same type of MS and people without MS.
These systems will help with tailored management of MS. Associate Professor Wang and Professor Barnett aim make these systems available for the wider research community to use, through existing research databases.
This includes the MSBase Imaging Repository (MSBIR), a unique “bank” of MRI scans of people with MS that is linked to clinical information through the MSBase registry. With their industry partner, Sydney Neuroimaging Analysis Centre, the team will translate these AI analyses of MRI into tools that can be used in the clinic to help measure and manage MS progression.
For this goal, Associate Professor Wang and Professor Barnett will conduct a ‘virtual clinical trial’, using data from MSBase and MSBIR, to better understand the effectiveness of current MS treatments on disease progression in a real-world setting.
Since the start of the project, Associate Professor Wang and Professor Barnett have focused on developing AI tools to detect “silent” MS progression. These are the changes that occur in disease progression without clinical symptoms, that standard MRI reporting often misses. Their team has successfully built AI models to accurately measure two critical signs of silent MS worsening: the slow expansion of existing MS lesions and the tissue loss in the neck (cervical) region of the spinal cord.
To help doctors better understand these advanced MRI measures, the researchers also provide comparisons for reference. Instead of just giving fixed numbers, the result is compared with a large group of people with a similar MS profile, helping them see whether a person’s MS is stable or getting worse.
The team also created a new AI assistant called “NeuroRAP,” which uses both imaging and clinical data to help predict how a person’s MS might progress over the next two years. The work on this predictive tool was presented at a major international MRI conference in Cape Town, of the International Society for Magnetic Resonance in Medicine 2026.
Finally, the team has started sharing these advanced analysis methods with the MS research community through MSBase/MSBIR.
All the data for this study has been collected. Associate Professor Wang and Professor Barnett expect more results within the coming year to assist in personalised disease management for people living with MS.
Ma, Y., Wang, D., Liu, P., Masters, L., Barnett, M., Cai, W., & Wang, C. (2024). Symmetry awareness encoded deep learning framework for brain imaging analysis. Medical Image Computing and Computer Assisted Intervention– MICCAI 2024.
Last updated 31 March 2026
$750,000
2024
3 years
Current project
