Research Article
Characterisation of Breathing and Physical Activity Patterns in the General Population Using the Wearable Respeck Monitor
@INPROCEEDINGS{10.1007/978-3-030-34833-5_6, author={D. Arvind and D. Fischer and C. Bates and S. Kinra}, title={Characterisation of Breathing and Physical Activity Patterns in the General Population Using the Wearable Respeck Monitor}, proceedings={Body Area Networks: Smart IoT and Big Data for Intelligent Health Management. 14th EAI International Conference, BODYNETS 2019, Florence, Italy, October 2-3, 2019, Proceedings}, proceedings_a={BODYNETS}, year={2019}, month={11}, keywords={Wearable sensors Respiratory rate Physical activity}, doi={10.1007/978-3-030-34833-5_6} }- D. Arvind
D. Fischer
C. Bates
S. Kinra
Year: 2019
Characterisation of Breathing and Physical Activity Patterns in the General Population Using the Wearable Respeck Monitor
BODYNETS
Springer
DOI: 10.1007/978-3-030-34833-5_6
Abstract
Clinical trials employing manual processes for data collection and administering of questionnaires are time-consuming, expensive to run and result in noisy data. Wireless body-worn sensors coupled with mobile applications can be harnessed to automate the data collection process during clinical trials. This paper describes the use of the Respeck monitor, worn as a plaster on the chest, for characterising breathing and physical activity patterns in the general population during their normal everyday lives. Respeck data collected from 93 subjects for periods ranging between 24 to 72 h, amounting to a total of 106 days of continuous Respeck data. Analysis of the data revealed new insights, such as the respiratory rate levels dropped by 4.39 breaths per minute (BrPM) on average during sleeping periods, compared to the preceding day-time periods. This change is higher than typically reported levels when normally measured directly before the subjects fall asleep. Previous research in activity patterns in the general population were based on high-level activities logged using questionnaires. A method is presented for clustering simple, yet high-dimensional, activity patterns based on the Respeck data, by first extracting relevant features for each day. The results reveal four distinct groups in the cohort corresponding to different identifiable lifestyles: “Sedentary”, “Moderately active”, “Active walkers” and “Active movers”.