AUTHOR=Liang Zilu , Chapa-Martell Mario Alberto TITLE=A Multi-Level Classification Approach for Sleep Stage Prediction With Processed Data Derived From Consumer Wearable Activity Trackers JOURNAL=Frontiers in Digital Health VOLUME=Volume 3 - 2021 YEAR=2021 URL=https://www.frontiersin.org/journals/digital-health/articles/10.3389/fdgth.2021.665946 DOI=10.3389/fdgth.2021.665946 ISSN=2673-253X ABSTRACT=Consumer wearable activity trackers such as Fitbit are widely used in ubiquitous and longitudinal sleep monitoring in free-living environments. However, these devices are known to be inaccurate for measuring sleep stages. In this study, we develop and validate a novel approach that leverages the processed data readily available from consumer activity trackers (i.e., steps, heart rate, and sleep metrics) to predict sleep stages. The proposed approach adopts a selective correction strategy and consists of two levels of classifiers. The level-I classifier judges whether a Fitbit labelled sleep epoch is misclassified, and the level-II classifier re-classifies misclassfied epochs into one of the four sleep stages (i.e., light sleep, deep sleep, REM sleep and wakefulness). Best epoch-wise performance was achieved when support vector machine and gradient boosting decision tree (XGBoost) with up sampling were used respectively at the level-I and level-II classification. The model achieved an overall per-epoch accuracy of 0.731 (SD: 0.119), Cohen's Kappa of 0.433 (SD: 0.212), and multi-class Matthew's correlation coefficient (MMCC) of 0.451 (SD: 0.214). Regarding the total duration of individual sleep stage, the mean normalized absolute bias (MAB) of this model was 0.469, which is a 23.9% reduction against the proprietary Fitbit algorithm. The model that combines support vector machine and XGBoost with down sampling achieved sub-optimal per-epoch accuracy of 0.704 (SD: 0.097), Cohen's Kappa of 0.427 (SD: 0.178), and MMCC of 0.439 (SD: 0.180). The sub-optimal model obtained a MAB of 0.179, a significantly reduction of 71.0% compared to the proprietary Fitbit algorithm. We highlight the challenges in machine learning based sleep stage prediction with consumer wearables, and suggest directions for future research.