Related Experiment Video
Updated: Jul 1, 2025
07:24
Home-Based Monitor for Gait and Activity Analysis
Published on: August 8, 2019
6.7K
Modeling multiple sclerosis using mobile and wearable sensor data
Shkurta Gashi1,2, Pietro Oldrati3, Max Moebus4
1Department of Computer Science, ETH Zürich, Zürich, Switzerland. shkurta.gashi@ai.ethz.ch.
NPJ Digital Medicine
|March 12, 2024
Summary
Remote monitoring using mobile and wearable sensors can effectively track multiple sclerosis (MS) progression and patient status. This approach aids in early diagnosis, treatment management, and clinical trial identification for people with MS (PwMS).
Area of Science:
- Neurology
- Biomedical Engineering
- Digital Health
Background:
- Multiple sclerosis (MS) is a leading cause of non-traumatic disability in young adults, necessitating frequent monitoring for timely intervention.
- Current diagnostic and monitoring methods for MS rely on infrequent clinical visits, hindering early treatment access and disease progression management.
- There is a critical need for remote, frequent monitoring solutions to improve the management of multiple sclerosis.
Purpose of the Study:
- To investigate the feasibility and clinical utility of mobile and wearable sensor-derived features for monitoring multiple sclerosis (MS).
- To differentiate people with MS (PwMS) from healthy controls using behavioral markers derived from sensor data.
- To assess the ability of these features to recognize MS disability and fatigue levels.
Main Methods:
- Collected a dataset from 55 PwMS and 24 healthy controls over 489 days in free-living conditions.
- Utilized data from arm-worn devices (heart rate), smartphone applications (phone locks), patient health records (MS type), and validated questionnaires (fatigue levels).
- Formalized clinical knowledge to derive behavioral markers characterizing MS from sensor and self-report data.
Main Results:
- Demonstrated the feasibility of using features from mobile and wearable sensors for monitoring multiple sclerosis (MS).
- Showcased the potential of these digital markers to distinguish people with MS (PwMS) from healthy individuals.
- Indicated the capability of the approach to recognize varying levels of MS-related disability and fatigue.
Conclusions:
- Mobile and wearable sensors offer a viable solution for continuous, remote monitoring of MS in real-world settings.
- This technology can facilitate effective disease management, treatment evaluation, and patient identification for clinical trials.
- The findings support the integration of digital health tools into the routine care of individuals with multiple sclerosis.
