JMIR Publications

ABSTRACT

Background:

Fatigue is the most common symptom among Multiple Sclerosis (MS) patients and severely affects quality of life. It is driven by factors such as disease state, age, stress, and physical activity as shown by recent literature. However, we currently have little understanding of the causes of short-term changes in fatigue as observed throughout the day.

Objective:

This study aims to investigate how perceived fatigue can be predicted using biomarkers collected from an arm-worn wearable sensor for MS patients and a healthy control group at an unprecedented time resolution of more than five times per day.

Methods:

On average, during our two-week study, 74 participants (51 MS patients, 23 Control) reported their level of fatigue 51 times totaling more than 3700 data points. We use interpretable Generalized Additive Models to arrive at the set of variables with the highest explanatory power for perceived fatigue reported on a visual analog scale.

Results:

Perceived fatigue follows a daily upward trend. We find significant effects of increased physical activity, heart rate, sympathetic activity, and parasympathetic activity while awake and asleep that explain deviations from the daily upward trend. Dysfunction of the autonomic nervous system (ANS) changes how biomarkers relate to perceived fatigue.

Conclusions:

In summary, we show that changes in perceived fatigue throughout the day can be predicted using biomarkers, especially the ones reflecting autonomic nervous system activity. Demonstrating that our constructed models can explain up to 34% of the variance in perceived fatigue responses, we believe that our analysis opens up new research opportunities for fine-grained modeling of perceived fatigue using physiological signals passively collected with wearable sensors.