Bridging the Gap: Making MS Diagnosis and Care Equitable through AI-Driven MRI Analysis in the NHS

Multiple Sclerosis (MS) affects over 130,000 people in the UK and is the leading cause of non-traumatic disability in young adults. MRI is the central to diagnosis and monitoring, but reporting is slow, inconsistent, and reliant on subspecialist neuroradiologists, who are in critically short supply. National audits show over half of NHS MS centres experience delayed reporting, with widespread outsourcing that often requires rereview. These delays restrict timely access to disease modifying therapies (DMTs) and contribute to inequities in care, particularly for underserved populations. AINOSTICS has developed BR[AI]N MS, an AI-driven tool that automates MRI analysis. It integrates (i) preprocessing and harmonisation across scanners and protocols, (ii) lesion segmentation and classification of key biomarkers including central vein sign (CVS) and paramagnetic rim lesions (PRLs), and (iii) structured reporting. In pilot testing, BR[AI]N MS achieved >90% accuracy for lesion segmentation and biomarker classification, stable across scanner vendors and field strengths, and processes a scan in approximately 5 minutes. Crucially, it enables consistent application of the updated 2024 revised McDonald criteria (Published in September 2025), where CVS and PRLs are incorporated as core diagnostic features, making their reliable identification essential for accurate diagnosis and prognosis. This project will validate BR[AI]N MS on a large, diverse NHS dataset, benchmark performance against expert annotations, assess cost-effectiveness, and model improvements in geographical equity of access. Usability testing with radiologists and neurologists will ensure clinical readiness, with active involvement from patients and the public. The outputs will include a validated minimum marketable product (MMP) capable of PACS/EHR integration, evidence of clinical and economic value, and a CE application to support adoption across the NHS.