Personalised profiling to identify clinically relevant changes in tremor due to multiple sclerosis.

Area of Science:

• Neurology
• Biomedical Engineering
• Movement Disorders

Background:

• Growing interest in sensor-based upper limb tremor assessment for multiple sclerosis and movement disorders.
• Previous sensor-based methods lacked improvement over clinical observation due to poor test-retest repeatability.
• A key barrier to sensor-based tremor assessment is distinguishing random variability from clinically relevant symptom changes.

Purpose of the Study:

• To overcome the barrier of poor repeatability in sensor-based tremor assessment.
• To develop customized tremor change metrics to differentiate random variability from clinically relevant symptom changes.
• To compare newly proposed metrics against conventional clinical and sensor-based metrics in individuals with multiple sclerosis tremor.

Main Methods:

• Developed novel 'change in scale' tremor change metrics, customized to individual tremor characteristics.
• Compared proposed metrics against conventional clinical and sensor-based metrics in 24 individuals with multiple sclerosis tremor.
• Evaluated metrics using Spearman rank correlation with Fahn-Tolosa-Marin Tremor Rating Scale (FTMTRS) subscales for functional disability (FTMTRS B) and self-assessed impact (FTMTRS C).

Main Results:

• Newly proposed 'change in scale' metrics showed statistically significant correlations with changes in self-assessed tremor impact (max R²>0.5, p<0.05).
• Proposed metrics outperformed conventional metrics in correlations with changes in task-based functional performance (R²=0.25 vs. R²=0.15).
• The new metrics demonstrated improved sensitivity to change compared to conventional clinical observation.

Conclusions:

• The proposed metrics enhance sensor-based tremor assessment sensitivity, surpassing conventional visual observation.
• These findings suggest the main barrier to translational impact in sensor-based tremor assessment can be overcome.
• Refined sensor-based tremor assessments could improve personalized treatment selection and clinical trial efficiency through standardization and sensitivity.