Related Experiment Video
Updated: Jul 16, 2026
Multi-Modal Home Sleep Monitoring in Older Adults
Published on: January 26, 2019
Longitudinal Infant Sleep Monitoring Using a Sensor-Enabled Responsive Bassinet: A Population-Scale Feasibility Study
Savannah Gluck1, Teresa A Lillis2, Karthik Aroor3
1Mrs. T.H. Chan Division of Occupational Science and Occupational Therapy, University of Southern California, Los Angeles, CA 90089, USA.
Insights
This study shows how smart bassinets can track infant sleep patterns from birth to six months. The technology provides new insights into infant fussing and caregiver responses, improving sleep monitoring.
Area of Science:
- Infant Sleep Science
- Biomedical Engineering
- Data Science
Background:
- Sleep disturbances in infants are linked to negative outcomes for both infants and caregivers.
- Limited data exists on the longitudinal sleep development of infants from birth to six months.
- Traditional infant sleep monitoring methods have significant limitations.
Purpose of the Study:
- To assess the feasibility of using Internet of Things (IoT) bassinet data for infant sleep monitoring.
- To derive traditional and novel sleep metrics from aggregated activity logs.
- To characterize infant fussing and the impact of responsive bassinet features.
Main Methods:
- Utilized aggregated activity logs from a commercial IoT bassinet.
- Analyzed data from 26,187 infants.
- Employed a data-driven approach to define sleep/fussing periods and select appropriate nights for analysis.
Main Results:
- Successfully derived traditional sleep metrics (longest sleep stretch, total night sleep, sleep efficiency).
- Identified novel metrics for infant fussing and caregiver intervention delays.
- Demonstrated unprecedented resolution in longitudinal infant sleep data.
Conclusions:
- IoT bassinet technology offers a feasible method for detailed infant sleep and fussing monitoring.
- This approach provides a framework for descriptive and comparative sleep analysis.
- Sensor-enabled responsive technology can capture dyadic interactions not measured by traditional methods.
Abstract:
Sleep is crucial to infant development, and excessive sleep disturbances are associated with adverse outcomes for both infants and their caregivers. There is limited information on the longitudinal development of sleep (e.g., duration, fragmentation, etc.) from birth to 6 months of age. New technologies, which include real-time environmental sensing and responses, have the potential to overcome many of the traditional limitations on infant sleep monitoring. In this study, we demonstrate the feasibility of utilizing aggregated activity logs from a commercially available IoT (Internet of Things) bassinet to derive traditional sleep metrics (longest sleep stretch, total night sleep, and sleep efficiency), as well as novel metrics related to infant fussing and impacts of the bed's ability to deliver responsive motion and sound. A total of 26,187 infants (1000-8000 per night) were included in this analysis. A data-driven approach was utilized to define the temporal boundaries of each night, divide each night into periods of sleep and fussing, and identify appropriate nights for inclusion. The derived data provide, in unprecedented resolution, a detailed longitudinal view of infant sleep in this specific population. Our results generally align with previous studies of traditional sleep metrics; however, they also demonstrate a methodological framework for descriptive or comparative monitoring of sleep and soothing, and uniquely characterize dyadic interactions that are not well-captured by traditional metrics. For example, the bassinet's activity logs indicate not only the proportion of fussing episodes that are resolved without caregiver intervention (e.g., removal), but also reflect the delay between fussing and the need for caregiver intervention. Further evaluation of this sensor-enabled, responsive technology in relation to sleep and fussing is merited.
