Imperial College London
Alternate

ProfessorStuartCook

Faculty of Medicine, Institute of Clinical Sciences

Visiting Professor
 
 
 
//

Contact

 

+44 (0)20 3313 1346stuart.cook

 
 
//

Location

 

RF 16Sydney StreetRoyal Brompton Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Zhou:2022:10.2196/34669,
author = {Zhou, W and Chan, YE and Foo, CS and Zhang, J and Teo, JX and Davila, S and Huang, W and Yap, J and Cook, S and Tan, P and Chin, CW-L and Yeo, KK and Lim, WK and Krishnaswamy, P},
doi = {10.2196/34669},
journal = {Journal of Medical Internet Research},
title = {High-resolution digital phenotypes from consumer wearables and their applications in machine learning of cardiometabolic risk markers: cohort study},
url = {http://dx.doi.org/10.2196/34669},
volume = {24},
year = {2022}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - Background:Consumer-grade wearable devices enable detailed recordings of heart rate and step counts in free-living conditions. Recent studies have shown that summary statistics from these wearable recordings have potential uses for longitudinal monitoring of health and disease states. However, the relationship between higher resolution physiological dynamics from wearables and known markers of health and disease remains largely uncharacterized.Objective:We aimed to derive high-resolution digital phenotypes from observational wearable recordings and to examine their associations with modifiable and inherent markers of cardiometabolic disease risk.Methods:We introduced a principled framework to extract interpretable high-resolution phenotypes from wearable data recorded in free-living conditions. The proposed framework standardizes the handling of data irregularities; encodes contextual information regarding the underlying physiological state at any given time; and generates a set of 66 minimally redundant features across active, sedentary, and sleep states. We applied our approach to a multimodal data set, from the SingHEART study (NCT02791152), which comprises heart rate and step count time series from wearables, clinical screening profiles, and whole genome sequences from 692 healthy volunteers. We used machine learning to model nonlinear relationships between the high-resolution phenotypes on the one hand and clinical or genomic risk markers for blood pressure, lipid, weight and sugar abnormalities on the other. For each risk type, we performed model comparisons based on Brier scores to assess the predictive value of high-resolution features over and beyond typical baselines. We also qualitatively characterized the wearable phenotypes for participants who had actualized clinical events.Results:We found that the high-resolution features have higher predictive value than typical baselines for clinical markers of cardiometabolic disease risk: the best models based on
AU  - Zhou,W
AU  - Chan,YE
AU  - Foo,CS
AU  - Zhang,J
AU  - Teo,JX
AU  - Davila,S
AU  - Huang,W
AU  - Yap,J
AU  - Cook,S
AU  - Tan,P
AU  - Chin,CW-L
AU  - Yeo,KK
AU  - Lim,WK
AU  - Krishnaswamy,P
DO  - 10.2196/34669
PY  - 2022///
SN  - 1438-8871
TI  - High-resolution digital phenotypes from consumer wearables and their applications in machine learning of cardiometabolic risk markers: cohort study
T2  - Journal of Medical Internet Research
UR  - http://dx.doi.org/10.2196/34669
UR  - https://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=PARTNER_APP&SrcAuth=LinksAMR&KeyUT=WOS:000966416900002&DestLinkType=FullRecord&DestApp=ALL_WOS&UsrCustomerID=a2bf6146997ec60c407a63945d4e92bb
UR  - https://www.jmir.org/2022/7/e34669
UR  - http://hdl.handle.net/10044/1/107616
VL  - 24
ER  -
Download