Wearable sensors may help identify patients with multiple sclerosis (MS) who are at increased risk of increasing limitations and brain atrophy. The researchers found that changes in daily activity patterns are associated with a greater likelihood of disease progression.
People whose movement patterns changed more significantly during the study period were more likely to experience an increase in limitations and loss of brain volume than participants whose patterns remained stable. The study was described by the American Academy of Neurology and published in the scientific journal Neurology.
Continuous monitoring of activity
The study followed 238 people with MS, with an average age of 55. The participants had been living with the disease for an average of 13 years. At the start of the study, they had an average score of 3 on the Expanded Disability Status Scale (EDSS), which indicates limited walking difficulties but moderate limitations in other neurological functions.
The participants wore a wearable device on their wrist for two weeks every three months. These sensors recorded various aspects of physical activity 24 hours a day, such as light, moderate, and vigorous exercise, time spent sitting or being inactive, and sleep-wake patterns.
In addition, neurological tests were performed every six months to measure changes in the level of limitations. MRI scans of the brain were performed at the start of the study and after two years to analyze changes in brain volume.
Decrease in activity as a possible signal
During the study period, 120 participants showed signs of disease progression. In particular, a decrease in physical activity during the day was found to be associated with a greater risk of deterioration. The researchers found that participants with less activity in the first half of the day were about 20 percent more likely to experience disease progression than those whose activity remained stable.
A decrease in activity in the morning, particularly between 8:00 and 10:00 a.m., was also associated with loss of brain volume. Each standard deviation decrease in activity was associated with a 0.18 percent decrease in total brain volume, a 0.34 percent loss in deep gray matter, and a 0.35 percent volume reduction in the thalamus.
Potential for early detection
According to Kathryn C. Fitzgerald of Johns Hopkins University, relatively simple wearables may help detect subtle changes in the disease earlier. “Identifying patients at risk for disease progression early is essential to reducing long-term disability, but current tests for measuring MS disability are not designed to detect small changes,” Fitzgerald says. “Using a relatively inexpensive and accessible wrist-worn device could help us identify early changes in the disease.”
The researchers emphasize that the study only shows a correlation between changing activity patterns and disease progression, and does not prove that reduced activity exacerbates the disease.
Further studies needed
Although the results are promising, the researchers say there are limitations. For example, there was no control group without MS, making it more difficult to determine the extent to which changes in activity are related to normal aging. In addition, the participants were relatively older and already had a certain degree of limitations, which means that the results may not be fully applicable to younger patients or people with milder forms of MS.
Nevertheless, the findings suggest that wearable sensors could play a role in the future in detecting disease progression at an earlier stage and possibly also in research into new treatments for MS.
AI predicts MS progression
Last year, Swedish researchers developed an AI model that can determine with approximately 90 percent certainty which form of multiple sclerosis (MS) a patient has. The model helps to earlier recognize when relapsing-remitting MS (RRMS) transitions to secondary progressive MS (SPMS), a transition that currently takes an average of three years to diagnose. This is important because both forms require different treatments.
The AI model analyzes clinical data from more than 22,000 patients in the Swedish MS registry, including neurological tests, MRI scans, and treatment information collected during regular consultations. By recognizing patterns in this data, the system can determine which stage of the disease a patient is in. In addition, the model also indicates how confident it is in each assessment, allowing doctors to gauge the reliability of the AI analysis.
