
A new scientific commentary in *Brain* by Tobias Brummer and Vinzenz Fleischer discusses our recent MRI and blood marker study as part of a wider change in multiple sclerosis (MS) research. The editorial argues that labels such as relapsing-remitting or progressive MS remain useful, but they are based mainly on symptoms seen over time. They do not always capture the biological processes happening in the brain and nervous system.

The commentary describes a move toward a more biological framework for MS: one that combines MRI, blood markers such as serum neurofilament light chain, and machine learning to identify patterns of tissue damage and nerve injury. In plain terms, this means asking not only what symptoms a person has, but what kind of disease activity appears to be unfolding underneath. This could eventually help clinicians estimate prognosis earlier and choose treatments in a more targeted way.

The authors also note that more work is needed before this approach can become routine care. Future studies will need to test these models in more diverse real-world MS populations and connect them to practical clinical tools. Even so, the editorial frames the MRI-serum study as a step toward a broader shift in neurology: moving from symptom-based categories toward molecular and biological descriptions of disease. Read the commentary in *Brain*: [Beyond clinical labels: a molecular-structural framework for multiple sclerosis subtyping](https://doi.org/10.1093/brain/awaf400).


Brain editorial highlights shift toward biological MS subtyping


A new open-access study in *Brain*, led by **Queen Square Analytics** in collaboration with academic and industry partners, shows that combining MRI scans with a blood marker called serum neurofilament light chain can reveal different biological patterns of multiple sclerosis (MS). This matters because MS can look very different from one person to another, and traditional labels such as relapsing or progressive MS do not always explain what is happening underneath.

The team used machine learning to analyse MRI measures alongside serum neurofilament light chain, a marker that can rise when nerve cells are injured. The model found two broad MS patterns: one where the blood marker became abnormal early alongside MRI signs of active injury, and another where brain tissue changes appeared earlier and the blood marker rose later. These patterns were tested in separate groups of people with MS, including people closer to diagnosis.

For people living with MS, the long-term goal is more personalised care: better ways to understand the likely course of disease and choose treatments that match the biology of each person. The findings suggest that using MRI and blood tests together may give a clearer picture than MRI alone, and could support future tools for predicting disease activity and treatment response. Read the paper in *Brain*: [Combined magnetic resonance imaging and serum analysis reveals distinct multiple sclerosis types](https://doi.org/10.1093/brain/awaf331).


MRI and blood marker study reveals distinct MS types


A new *Nature Communications* paper introduces **MindGlide**, an AI model that enables robust quantification of key multiple sclerosis (MS) MRI biomarkers—**lesion burden and regional brain volumes—from a single routine MRI contrast**. By removing the usual dependency on multi-contrast research protocols, MindGlide is designed to turn “old” and heterogeneous hospital MRI archives into analysable datasets, helping bring quantitative MRI closer to real-world care and research. ([Nature](https://www.nature.com/articles/s41467-025-58274-8 "Enabling new insights from old scans by repurposing clinical MRI ..."))

MindGlide was trained on **4,247** MS scans from **2,934** patients across **592** scanners, then externally validated on **14,952** scans from **1,001** patients—including large progressive MS trial datasets and a routine-care cohort. The model **outperformed comparator tools** against expert-labelled lesion volumes and, crucially, could **detect treatment effects** on lesion accrual and grey-matter tissue loss in clinical trials—even using contrasts not typically used for those outcomes (e.g., T2-weighted scans). ([Nature](https://www.nature.com/articles/s41467-025-58274-8 "Enabling new insights from old scans by repurposing clinical MRI ..."))

MindGlide is **publicly released** with code, trained models, and a containerised environment, supporting reproducible deployment and community reuse. For MS-PINPOINT, this is a practical step toward scalable, clinic-compatible analytics: extracting meaningful biomarkers from routine scans can reduce barriers to real-world evidence studies and broaden the impact of existing MRI data collected over decades. ([GitHub](https://github.com/MS-PINPOINT/mindGlide "MS-PINPOINT/mindGlide: Brain Segmentation with MONAI ... - GitHub"))


News

Brain editorial highlights shift toward biological MS subtyping

MRI and blood marker study reveals distinct MS types

MindGlide unlocks routine MRI archives to accelerate MS research