- Currently, doctors categorize and treat cases of multiple sclerosis (MS) according to the way an individual’s symptoms progress.
- In a new study, researchers used artificial intelligence (AI) to analyze the brain scans of thousands of people with MS to learn more about the disease.
- The analysis revealed three MS subtypes based on brain abnormalities associated with MS.
- The researchers hope that the deeper understanding of MS that these new subtypes provide will help doctors more effectively target treatment.
MS is an autoimmune condition in which the immune system attacks the myelin sheath that covers and protects the nerves in the brain and spinal cord.
This chronic nervous system disease disrupts the flow of electrical signals between the brain and body, which often has life-changing consequences. Its progression is unpredictable, and treatment is difficult.
Currently, which one of the four symptom-based categories of MS a person’s condition falls under will determine their treatment options. However, this approach goes only so far.
“It does not directly rely on the underlying biology of the disease and, therefore, cannot assist doctors in choosing the right treatment for the right patients,” says Dr. Arman Eshaghi of the Queen Square Institute of Neurology at University College London (UCL).
Dr. Eshaghi is the lead author of a new study, which the researchers carried out to develop a better understanding of the physiological mechanisms underpinning clinical symptoms of MS.
Using AI, the study has identified three new subtypes of MS that may help doctors more effectively target treatment.
The study appears in the journal Nature Communications.
The researchers identified the new MS subtypes after performing MRI scans of the brains of 6,322 people with MS. Key to the discovery was the unsupervised analysis of the scans by a UCL-developed AI program called “SuStaIn,” which stands for “Subtype and Staging Inference.”
AI and machine learning are especially adept at seeing patterns in data that may be too subtle for humans to detect. “Here,” says Dr. Eshaghi, “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.”
The three new subtypes are “cortex-led,” “normal-appearing white matter-led,” and “lesion-led,” with the names describing how they first present as brain abnormalities.
To validate their findings, the researchers switched SuStaIn from analysis mode to detection mode and fed it a separate dataset of 3,068 additional brain MRI scans from people with MS. The brain abnormalities in this second group confirmed the existence of the three subtypes.
Dr. Eshaghi notes that following the identification of the subtypes, the team “did a further retrospective analysis of patient records to see how people with the newly identified MS subtypes responded to various treatments.”
While cautioning that further clinical studies are necessary, Dr. Eshaghi reports:
“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 toward predicting individual responses to therapies.”
Senior author Prof. Olga Ciccarelli notes that AI analysis of brain scans may be only the beginning: “The method used to classify MS is currently focused on imaging changes only; we are extending the approach to including other clinical information.”
Given the stubborn and still somewhat mysterious nature of MS, Prof. Ciccarelli says that the team’s study offers new hope for people dealing with this potentially debilitating disease:
“This exciting field of research will lead to an individual definition of MS course and individual prediction of treatment response in MS using AI.”
Prof. Ciccarelli foresees a time when it will be possible for doctors to “select the right treatment for the right patient at the right time.”
As it is, the study’s findings may be of use in selecting individuals with MS for research, say the study’s authors. In particular, the subtypes may allow the grouping of participants for targeted clinical treatment trials.
As another senior study author, Prof. Alan Thompson of UCL, notes, “We are aware of the limitations of the current descriptors of MS, which can be less than clear when applied to prescribing treatment.”
He notes that the discovery of the new MS subtypes may solve this problem:
“Now, with the help of AI and large datasets, we have made the first step toward a better understanding of the underlying disease mechanisms, which may inform our current clinical classification. This is a fantastic achievement and has the potential to be a real game-changer, informing both disease evolution and selection of patients for clinical trials.”