Assessment of Fatigue Using Wearable Sensors: A Pilot Study

Abstract

<b><i>Background:</i></b> Fatigue is a broad, multifactorial concept encompassing feelings of reduced physical and mental energy levels. Fatigue strongly impacts patient health-related quality of life across a huge range of conditions, yet, to date, tools available to understand fatigue are severely limited. <b><i>Methods:</i></b> After using a recurrent neural network-based algorithm to impute missing time series data form a multisensor wearable device, we compared supervised and unsupervised machine learning approaches to gain insights on the relationship between self-reported non-pathological fatigue and multimodal sensor data. <b><i>Results:</i></b> A total of 27 healthy subjects and 405 recording days were analyzed. Recorded data included continuous multimodal wearable sensor time series on physical activity, vital signs, and other physiological parameters, and daily questionnaires on fatigue. The best results were obtained when using the causal convolutional neural network model for unsupervised representation learning of multivariate sensor data, and random forest as a classifier trained on subject-reported physical fatigue labels (weighted precision of 0.70 ± 0.03 and recall of 0.73 ± 0.03). When using manually engineered features on sensor data to train our random forest (weighted precision of 0.70 ± 0.05 and recall of 0.72 ± 0.01), both physical activity (energy expenditure, activity counts, and steps) and vital signs (heart rate, heart rate variability, and respiratory rate) were important parameters to measure. Furthermore, vital signs contributed the most as top features for predicting mental fatigue compared to physical ones. These results support the idea that fatigue is a highly multimodal concept. Analysis of clusters from sensor data highlighted a digital phenotype indicating the presence of fatigue (95% of observations) characterized by a high intensity of physical activity. Mental fatigue followed similar trends but was less predictable. Potential future directions could focus on anomaly detection assuming longer individual monitoring periods. <b><i>Conclusion:</i></b> Taken together, these results are the first demonstration that multimodal digital data can be used to inform, quantify, and augment subjectively captured non-pathological fatigue measures.

Evaluation, Acceptance, and Qualification of Digital Measures: From Proof of Concept to Endpoint

“…One possible next step is to replace, prompt, or augment COA items and subscales with digital measures derived from PGHD. It has been shown that PGHD can be used to accurately predict subjective mobility [47], stress [61, 62], and fatigue [63] outcomes, demonstrating the potential of this idea.…”

The Challenging Path to Developing a Mobile Health Device for Epilepsy: The Current Landscape and Where We Go From Here

“…A phase 2 EMU study, using a retrospective cross-validation approach, demonstrated that fusing accelerometry signals with electrodermal activity (EDA) recorded from a wrist-worn device, resulted in a sensitivity of 95% for GTCS and a false positive rate of 0.2 per day (99). Most recently, a phase 3 EMU study applied pre-defined cut-off points to data obtained in real-time, and showed that when heart rate changes and oximetry endpoints are combined, the sensitivity is highest, and lowest when the parameters are used alone (117).…”

Digital Phenotyping in Livestock Farming

“…When phenotyping data are collected, they are normally observed in association with a certain target, which leaves blind spots for influence by variables that were not apprehended. This event is well explained by a study that sought to determine the cause of fatigue in humans, only to find that the experience of fatigue hinged on diverse phenotypes that were both environmentally and biologically influenced [28]. The method of phenotyping with a specific focus can also influence inaccurate conclusions and actions that lead to trade-offs.…”