Fitness trackers and smartphones offer insights into multiple sclerosis management

Multiple sclerosis (MS) is an insidious disease. Patients suffer because their immune system attacks their own nerve fibers, disrupting the transmission of nerve signals. MS patients experience impairments in motor function and sensory perception in a variety of ways, ranging from mild to severe. These disorders interfere with daily life and reduce the overall quality of life. The symptoms and progression of the disease vary from person to person, and so do the treatments. To monitor the progression of the disease and be able to recommend effective treatments, doctors regularly ask patients about symptoms such as fatigue.

disappear from memory

Patients are therefore faced with the difficult task of having to provide information about their health status and what they were able to do based on their memory over the past few weeks or even months. Data collected this way can be inaccurate and incomplete because patients may remember details incorrectly or adjust their responses to society's expectations. And these reactions can be mismanaged because they have a major impact on how disease progression is recorded.

“Physicians would benefit from access to reliable, frequent, and long-term measurements of patient health parameters that provide an accurate and comprehensive picture of a patient's health status,” explains Shukrta Gashi. .

She is the lead author of the new study, a postdoctoral researcher in the group led by Professor Christian Holz and Professor Gunnar Retsch from the ETH Department of Computer Science, and a fellow at the ETH AI Center.

In collaboration with colleagues at ETH Zurich, University Hospital Zurich, and the University of Zurich, Dr. Gashi has shown that fitness trackers and smartphones can provide this kind of reliable long-term data with high temporal resolution.Their research was published in the journal NPJ Digital Medicine.

Digital marker for MS

The researchers recruited a group of volunteers (55 people with MS and an additional 24 as control subjects) and provided each with an armband for fitness tracking. Researchers collected data from these wearable devices and the participants' smartphones over the course of two weeks. We then performed statistical tests and machine learning analysis on this data to identify reliable and clinically useful information.

What proved particularly meaningful was data on physical activity and heart rate collected from participants' wearable devices. It was found that the higher the severity of illness and fatigue level of participants, the lower their physical activity and heart rate variability. Compared to the control group, MS patients took fewer steps per day, had lower overall levels of physical activity, and had more consistent heartbeat intervals.

The frequency with which people use their smartphones also provides important information about the severity of illness and fatigue levels. The less frequently study participants used their smartphones, the greater the level of disability and the more severe the level of fatigue. Researchers gained insight into motor function through a game-like smartphone test. Developed several years ago at ETH, this test requires users to tap the screen as quickly as possible in order to move a virtual person as quickly as possible. By monitoring how fast a person taps and how that tap frequency changes over time, researchers can draw conclusions about a person's athletic performance and physical fatigue.

Taken together, the combination of data from fitness trackers and smartphones allows us to distinguish between healthy participants and those with MS with high accuracy. Combining information related to several aspects of the disease, such as physiological, behavioral, motor performance, and sleep information, is of great importance for more effective and accurate monitoring of the disease. ”

Shortgashi, first author

reliable approach

This new approach provides MS patients with an easy way to collect reliable and clinically useful long-term data while they go about their daily lives. Researchers hope this type of data could lead to better treatments and more effective disease management techniques. More comprehensive, accurate, and reliable data can help experts make better decisions and, in some cases, even suggest effective treatments sooner than before. Furthermore, by evaluating this patient data, experts can verify the effectiveness of different treatments.

The researchers are now making their dataset available to other scientists. They also point out that larger studies and more data are needed to develop reliable and generalizable models for automated assessment. In the future, such models could significantly improve the lives of MS patients thanks to data from fitness trackers and smartphones.