Movement-Based Sleep Estimation Using a Smart Mattress: Adaptation of Cole–Kripke Algorithm and Comparison with a Multimodal Wearable

Multimodal consumer wearables such as Garmin smartwatches combine movement with physiological signals to estimate sleep, whereas bed-based sensing largely relies on movement alone. How far movement alone, sensed from the bed, can go in estimating sleep is unclear. We adapted the Cole–Kripke algorithm to the movement signal of a WhizPad, a smart mattress with a 30-zone pressure array, using the original coefficients and one device-specific scale factor. Estimates were compared with those of a Garmin smartwatch, a proprietary comparator rather than a reference standard, across 28 paired nights in four participants. Epoch-level agreement between the two systems was high for sleep (98.08%) but substantially poorer for wake (30.76%; Cohen’s κ = 0.346). Over each system’s own sleep period, pooled wake after sleep onset (WASO) was 7.3 min lower for the WhizPad and pooled total sleep time (TST) 37.2 min higher, largely through a longer sleep period. The mattress registers bed entry, which Garmin does not export, allowing an exploratory, device-derived estimate of sleep latency that was not independently validated. These findings are descriptive and exploratory, and agreement does not establish the accuracy of either system as no polysomnography (PSG) reference was used. Physiological sensing in the bed and, ultimately, PSG validation are directions for future work.

Authors

Institutions

Publication Details

Journal
Sensors
Published
2026-11-12
DOI
https://doi.org/10.3390/s26227137
Primary Topic
Sleep and related disorders
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Movement-Based Sleep Estimation Using a Smart Mattress: Adaptation of Cole–Kripke Algorithm and Comparison with a Multimodal Wearable

Binquan Zhang, Yeh-Liang Hsu, Ankush Banerjee, He-Yuan Chen
Sensors
Sleep and related disorders
article

Movement-Based Sleep Estimation Using a Smart Mattress: Adaptation of Cole–Kripke Algorithm and Comparison with a Multimodal Wearable

Binquan Zhang, Yeh-Liang Hsu, Ankush Banerjee, He-Yuan Chen
article en

Abstract

Multimodal consumer wearables such as Garmin smartwatches combine movement with physiological signals to estimate sleep, whereas bed-based sensing largely relies on movement alone. How far movement alone, sensed from the bed, can go in estimating sleep is unclear. We adapted the Cole–Kripke algorithm to the movement signal of a WhizPad, a smart mattress with a 30-zone pressure array, using the original coefficients and one device-specific scale factor. Estimates were compared with those of a Garmin smartwatch, a proprietary comparator rather than a reference standard, across 28 paired nights in four participants. Epoch-level agreement between the two systems was high for sleep (98.08%) but substantially poorer for wake (30.76%; Cohen’s κ = 0.346). Over each system’s own sleep period, pooled wake after sleep onset (WASO) was 7.3 min lower for the WhizPad and pooled total sleep time (TST) 37.2 min higher, largely through a longer sleep period. The mattress registers bed entry, which Garmin does not export, allowing an exploratory, device-derived estimate of sleep latency that was not independently validated. These findings are descriptive and exploratory, and agreement does not establish the accuracy of either system as no polysomnography (PSG) reference was used. Physiological sensing in the bed and, ultimately, PSG validation are directions for future work.

SensorsVol. 26(22)
Yuan Ze University (TW)
Sleep and related disorders
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon