TILES-2018 Sleep Benchmark Dataset: A Longitudinal Wearable Sleep Data Set of Hospital Workers for Modeling and Understanding Sleep Behaviors

2025-07-04

AI Summary

This study addresses the challenge of modeling sleep behavior in real-world settings. We introduce the first large-scale, longitudinal wearable sleep dataset specifically designed for healthcare workers, comprising 139 participants observed over 10 weeks with over 6,000 sleep episodes. The dataset integrates multimodal, synchronized measurements: wrist-worn accelerometer and heart rate signals, alongside self-reported sleep diaries and validated psychological scales (e.g., anxiety and perceived stress). Its key innovation lies in the first systematic collection and public release of continuous, ecologically valid sleep data from healthcare professionals—filling a critical gap in high-quality, naturalistic sleep datasets for this high-risk population. Leveraging this resource, we establish three benchmark tasks: sleep-stage classification, prediction of self-reported sleep quality, and demographic attribute identification. Experimental results demonstrate strong performance and generalizability, validating the dataset’s utility for personalized sleep modeling and health-status association analysis.

Technology Category

• Application Domains: Humanities & Computational Social Science
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Cognitive Modeling & Cognitive Systems: Affective Computing

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• Security and Privacy: Large-scale security measurements
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience

Abstract

Sleep is important for everyday functioning, overall well-being, and quality of life. Recent advances in wearable sensing technology have enabled continuous, noninvasive, and cost-effective monitoring of sleep patterns in real-world natural living settings. Wrist-worn devices, in particular, are capable of tracking sleep patterns using accelerometers and heart rate sensors. To support sleep research in naturalistic environments using wearable sensors, we introduce the TILES-2018 Sleep Benchmark dataset, which we make publicly available to the research community. This dataset was collected over a 10-week period from 139 hospital employees and includes over 6,000 unique sleep recordings, alongside self-reported survey data from each participant, which includes sleep quality, stress, and anxiety among other measurements. We present in-depth analyses of sleep patterns by combining the TILES-2018 Sleep Benchmark dataset with a previously released dataset (TILES-2018), which follows a similar study protocol. Our analyses include sleep duration, sleep stages, and sleep diaries. Moreover, we report machine learning benchmarks using this dataset as a testbed for tasks including sleep stage classification, prediction of self-reported sleep quality, and classifying demographics. Overall, this dataset provides a valuable resource for advancing foundational studies in sleep behavior modeling.

Problem

• Modeling sleep behaviors using wearable sensor data
• Predicting sleep quality and stress from sensor data
• Classifying sleep stages and demographics via machine learning

Innovation

• Longitudinal wearable sleep data collection
• Combining accelerometer and heart rate sensors
• Machine learning benchmarks for sleep analysis