A machine learning algorithm can use data from brain magnetic resonance imaging (MRI) to identify subtypes of multiple sclerosis (MS).  That’s according to a report published recently in Nature Communications by a team of researchers led by Arman Eshaghi, MD.

For the study, the researchers applied unsupervised machine learning to brain MRI scans from previously published studies to determine if there were any still-unidentified patterns in brain images that could better guide treatment choice and identify patients who would best respond to a particular therapy.

The training dataset of 6,322 MS patients was used to define MRI-based subtypes, and an independent validation cohort consisted of 3,068 patients.

Based on the earliest abnormalities, the researchers defined MS subtypes as cortex-led, normal-appearing white matter-led, and lesion-led. The highest risk for confirmed disability progression and the highest relapse rate were seen among people with the lesion-led subtype. However, people with the lesion-led MS subtype also showed positive treatment response in selected clinical trials.

"MS is unpredictable and different for everyone, and we know one of our community's main concerns is how their condition might develop," Clare Walton, Ph.D., from the MS Society, said with the report’s release. "Having an MRI-based model to help predict future progression and tailor your treatment plan accordingly could be hugely reassuring to those affected. These findings also provide valuable insight into what drives progression in MS, which is crucial to finding new treatments.”

According to the report, MS affects over 2.8 million people globally, and is currently classified into four phenotypes, which are defined as either relapsing or progressive. Patients are categorized by a mixture of clinical observations, assisted by MRI brain images, and patients’ symptoms. These observations guide the timing and choice of treatment.

“Here, we used artificial intelligence and asked the question: can AI find MS subtypes that follow a certain pattern on brain images? Our AI has uncovered three data-driven MS subtypes that are defined by pathological abnormalities seen on brain images,” explained Dr. Eshaghi.

“We did a further retrospective analysis of patient records to see how people with the newly identified MS subtypes responded to various treatments. 

“While further clinical studies are needed, there was a clear difference, by subtype, in patients’ response to different treatments and in accumulation of disability over time. This is an important step towards predicting individual responses to therapies.”

Researchers say the findings suggest that MRI-based subtypes predict MS disability progression and response to treatment and can now be used to define groups of patients in interventional trials. Prospective research with clinical trials is required as the next step to confirm these findings.