Bayesian inference federated learning for heart rate prediction

L. Fang; X. Liu; X. Su; J. Ye; S. Dobson; P. Hui; S. Tarkoma

doi:10.1007/978-3-030-70569-5_8

Bayesian inference federated learning for heart rate prediction

L. Fang, X. Liu, X. Su, J. Ye, S. Dobson, P. Hui, S. Tarkoma

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The advances of sensing and computing technologies pave the way to develop novel applications and services for wearable devices. For example, wearable devices measure heart rate, which accurately reflects the intensity of physical exercise. Therefore, heart rate prediction from wearable devices benefits users with optimization of the training process. Conventionally, Cloud collects user data from wearable devices and conducts inference. However, this paradigm introduces significant privacy concerns. Federated learning is an emerging paradigm that enhances user privacy by remaining the majority of personal data on users’ devices. In this paper, we propose a statistically sound, Bayesian inference federated learning for heart rate prediction with autoregression with exogenous variable (ARX) model. The proposed privacy-preserving method achieves accurate and robust heart rate prediction. To validate our method, we conduct extensive experiments with real-world outdoor running exercise data collected from wearable devices.

Original language	English
Title of host publication	Wireless Mobile Communication and Healthcare
Subtitle of host publication	9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings
Editors	Juan Ye, Michael J. O'Grady, Gabriele Civitarese, Kristina Yordanova
Place of Publication	Cham
Publisher	Springer
Pages	116-130
Number of pages	15
ISBN (Electronic)	9783030705695
ISBN (Print)	9783030705688
DOIs	https://doi.org/10.1007/978-3-030-70569-5_8
Publication status	Published - 2021

Publication series

Name	Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering
Volume	362 LNICST
ISSN (Print)	1867-8211
ISSN (Electronic)	1867-822X

Keywords

Bayesian inference
Federated learning
Heart rate prediction
Wearable computing

Access to Document

10.1007/978-3-030-70569-5_8

Bayesian_inference_federated_learning_for_heartrate_prediction
Copyright © 2021 ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering. This work has been made available online in accordance with publisher policies or with permission. Permission for further reuse of this content should be sought from the publisher or the rights holder. This is the author created accepted manuscript following peer review and may differ slightly from the final published version. The final published version of this work is available at https://doi.org/10.1007/978-3-030-70569-5_8.
Accepted author manuscript, 412 KB

Science of Sensor System Software: Science of Sensor System Software
Dobson, S. A. (PI)
EPSRC
1/01/16 → 31/12/22
Project: Standard

Cite this

Fang, L., Liu, X., Su, X., Ye, J., Dobson, S., Hui, P., & Tarkoma, S. (2021). Bayesian inference federated learning for heart rate prediction. In J. Ye, M. J. O'Grady, G. Civitarese, & K. Yordanova (Eds.), Wireless Mobile Communication and Healthcare: 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings (pp. 116-130). (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering; Vol. 362 LNICST). Springer. https://doi.org/10.1007/978-3-030-70569-5_8

Fang, L. ; Liu, X. ; Su, X. et al. / Bayesian inference federated learning for heart rate prediction. Wireless Mobile Communication and Healthcare: 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings. editor / Juan Ye ; Michael J. O'Grady ; Gabriele Civitarese ; Kristina Yordanova. Cham : Springer, 2021. pp. 116-130 (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering).

@inproceedings{54d8fcabdc8c40ccb2dfb1a56bf9f60d,

title = "Bayesian inference federated learning for heart rate prediction",

abstract = "The advances of sensing and computing technologies pave the way to develop novel applications and services for wearable devices. For example, wearable devices measure heart rate, which accurately reflects the intensity of physical exercise. Therefore, heart rate prediction from wearable devices benefits users with optimization of the training process. Conventionally, Cloud collects user data from wearable devices and conducts inference. However, this paradigm introduces significant privacy concerns. Federated learning is an emerging paradigm that enhances user privacy by remaining the majority of personal data on users{\textquoteright} devices. In this paper, we propose a statistically sound, Bayesian inference federated learning for heart rate prediction with autoregression with exogenous variable (ARX) model. The proposed privacy-preserving method achieves accurate and robust heart rate prediction. To validate our method, we conduct extensive experiments with real-world outdoor running exercise data collected from wearable devices. ",

keywords = "Bayesian inference, Federated learning, Heart rate prediction, Wearable computing",

author = "L. Fang and X. Liu and X. Su and J. Ye and S. Dobson and P. Hui and S. Tarkoma",

note = "This work has been partially supported by the UK EPSRC under grant number EP/N007565/1, “Science of Sensor Systems Software”, and by Academy of Finland projects, grant number 325774, 3196669, 319670, 326305, and 325570.",

year = "2021",

doi = "10.1007/978-3-030-70569-5_8",

language = "English",

isbn = "9783030705688",

series = "Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering",

publisher = "Springer",

pages = "116--130",

editor = "Juan Ye and O'Grady, {Michael J.} and Gabriele Civitarese and Kristina Yordanova",

booktitle = "Wireless Mobile Communication and Healthcare",

address = "Netherlands",

}

Fang, L, Liu, X, Su, X, Ye, J , Dobson, S, Hui, P & Tarkoma, S 2021, Bayesian inference federated learning for heart rate prediction. in J Ye, MJ O'Grady, G Civitarese & K Yordanova (eds), Wireless Mobile Communication and Healthcare: 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings. Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering, vol. 362 LNICST, Springer, Cham, pp. 116-130. https://doi.org/10.1007/978-3-030-70569-5_8

Bayesian inference federated learning for heart rate prediction. / Fang, L.; Liu, X.; Su, X. et al.
Wireless Mobile Communication and Healthcare: 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings. ed. / Juan Ye; Michael J. O'Grady; Gabriele Civitarese; Kristina Yordanova. Cham: Springer, 2021. p. 116-130 (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering; Vol. 362 LNICST).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Bayesian inference federated learning for heart rate prediction

AU - Fang, L.

AU - Liu, X.

AU - Su, X.

AU - Ye, J.

AU - Dobson, S.

AU - Hui, P.

AU - Tarkoma, S.

N1 - This work has been partially supported by the UK EPSRC under grant number EP/N007565/1, “Science of Sensor Systems Software”, and by Academy of Finland projects, grant number 325774, 3196669, 319670, 326305, and 325570.

PY - 2021

Y1 - 2021

N2 - The advances of sensing and computing technologies pave the way to develop novel applications and services for wearable devices. For example, wearable devices measure heart rate, which accurately reflects the intensity of physical exercise. Therefore, heart rate prediction from wearable devices benefits users with optimization of the training process. Conventionally, Cloud collects user data from wearable devices and conducts inference. However, this paradigm introduces significant privacy concerns. Federated learning is an emerging paradigm that enhances user privacy by remaining the majority of personal data on users’ devices. In this paper, we propose a statistically sound, Bayesian inference federated learning for heart rate prediction with autoregression with exogenous variable (ARX) model. The proposed privacy-preserving method achieves accurate and robust heart rate prediction. To validate our method, we conduct extensive experiments with real-world outdoor running exercise data collected from wearable devices.

AB - The advances of sensing and computing technologies pave the way to develop novel applications and services for wearable devices. For example, wearable devices measure heart rate, which accurately reflects the intensity of physical exercise. Therefore, heart rate prediction from wearable devices benefits users with optimization of the training process. Conventionally, Cloud collects user data from wearable devices and conducts inference. However, this paradigm introduces significant privacy concerns. Federated learning is an emerging paradigm that enhances user privacy by remaining the majority of personal data on users’ devices. In this paper, we propose a statistically sound, Bayesian inference federated learning for heart rate prediction with autoregression with exogenous variable (ARX) model. The proposed privacy-preserving method achieves accurate and robust heart rate prediction. To validate our method, we conduct extensive experiments with real-world outdoor running exercise data collected from wearable devices.

KW - Bayesian inference

KW - Federated learning

KW - Heart rate prediction

KW - Wearable computing

U2 - 10.1007/978-3-030-70569-5_8

DO - 10.1007/978-3-030-70569-5_8

M3 - Conference contribution

SN - 9783030705688

T3 - Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering

SP - 116

EP - 130

BT - Wireless Mobile Communication and Healthcare

A2 - Ye, Juan

A2 - O'Grady, Michael J.

A2 - Civitarese, Gabriele

A2 - Yordanova, Kristina

PB - Springer

CY - Cham

ER -

Fang L, Liu X, Su X, Ye J , Dobson S, Hui P et al. Bayesian inference federated learning for heart rate prediction. In Ye J, O'Grady MJ, Civitarese G, Yordanova K, editors, Wireless Mobile Communication and Healthcare: 9th EAI International Conference, MobiHealth 2020, Virtual Event, November 19, 2020, Proceedings. Cham: Springer. 2021. p. 116-130. (Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering). Epub 2021 Feb 21. doi: 10.1007/978-3-030-70569-5_8

Bayesian inference federated learning for heart rate prediction

Abstract

Publication series

Keywords

Access to Document

Fingerprint

Projects

Science of Sensor System Software: Science of Sensor System Software

Cite this