Physiological Stress Level Estimation Based on Smartphone Logs

Naoki Yamamoto, Keiichi Ochiai, Akiya Inagaki, Yusuke Fukazawa, Masatoshi Kimoto, Kazuki Kiriu, Kouhei Kaminishi, Jun Ota, Tsukasa Okimura, Yuuri Terasawa, Takaki Maeda

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

Recently, inferring the state of people's mental health via passive mobile sensing has attracted significant attention. Previous studies have used the self-assessed stress levels as the ground truth; however, these are subjective measures. In this study, we use a physiologically-assessed stress metric to minimize the effect of participant subjectivity and further estimate it using behavioral features based on the smartphone usage logs. We initially requested the study participants (39 participants) to attach heart rate sensors for 8 hours per day and simultaneously collected continuous heart rate data and smartphone logs for 42 days. Further, we divided the participants into four types via clustering using the behavioral features derived from their smartphone sensor logs and trained each model via supervised learning using the heart rate data as the ground truth. Our results exhibit that the proposed method is more accurate (71%) as compared to the baseline method (54%). This demonstrates that physiologically-assessed stress levels can be estimated based on the implicit features that are gathered from the smartphone logs.

Original languageEnglish
Title of host publication2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9784907626341
DOIs
Publication statusPublished - 2019 Feb 26
Event11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018 - Auckland, New Zealand
Duration: 2018 Oct 52018 Oct 8

Publication series

Name2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018

Conference

Conference11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018
CountryNew Zealand
CityAuckland
Period18/10/518/10/8

Fingerprint

Smartphones
Supervised learning
Sensors
Health

Keywords

  • LF/HF
  • mental health
  • smartphone log
  • stress

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Yamamoto, N., Ochiai, K., Inagaki, A., Fukazawa, Y., Kimoto, M., Kiriu, K., ... Maeda, T. (2019). Physiological Stress Level Estimation Based on Smartphone Logs. In 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018 [8653590] (2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.23919/ICMU.2018.8653590

Physiological Stress Level Estimation Based on Smartphone Logs. / Yamamoto, Naoki; Ochiai, Keiichi; Inagaki, Akiya; Fukazawa, Yusuke; Kimoto, Masatoshi; Kiriu, Kazuki; Kaminishi, Kouhei; Ota, Jun; Okimura, Tsukasa; Terasawa, Yuuri; Maeda, Takaki.

2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018. Institute of Electrical and Electronics Engineers Inc., 2019. 8653590 (2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Yamamoto, N, Ochiai, K, Inagaki, A, Fukazawa, Y, Kimoto, M, Kiriu, K, Kaminishi, K, Ota, J, Okimura, T, Terasawa, Y & Maeda, T 2019, Physiological Stress Level Estimation Based on Smartphone Logs. in 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018., 8653590, 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018, Institute of Electrical and Electronics Engineers Inc., 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018, Auckland, New Zealand, 18/10/5. https://doi.org/10.23919/ICMU.2018.8653590
Yamamoto N, Ochiai K, Inagaki A, Fukazawa Y, Kimoto M, Kiriu K et al. Physiological Stress Level Estimation Based on Smartphone Logs. In 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018. Institute of Electrical and Electronics Engineers Inc. 2019. 8653590. (2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018). https://doi.org/10.23919/ICMU.2018.8653590
Yamamoto, Naoki ; Ochiai, Keiichi ; Inagaki, Akiya ; Fukazawa, Yusuke ; Kimoto, Masatoshi ; Kiriu, Kazuki ; Kaminishi, Kouhei ; Ota, Jun ; Okimura, Tsukasa ; Terasawa, Yuuri ; Maeda, Takaki. / Physiological Stress Level Estimation Based on Smartphone Logs. 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018. Institute of Electrical and Electronics Engineers Inc., 2019. (2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018).
@inproceedings{1324a88c1a8f4247902f707ca2de668c,
title = "Physiological Stress Level Estimation Based on Smartphone Logs",
abstract = "Recently, inferring the state of people's mental health via passive mobile sensing has attracted significant attention. Previous studies have used the self-assessed stress levels as the ground truth; however, these are subjective measures. In this study, we use a physiologically-assessed stress metric to minimize the effect of participant subjectivity and further estimate it using behavioral features based on the smartphone usage logs. We initially requested the study participants (39 participants) to attach heart rate sensors for 8 hours per day and simultaneously collected continuous heart rate data and smartphone logs for 42 days. Further, we divided the participants into four types via clustering using the behavioral features derived from their smartphone sensor logs and trained each model via supervised learning using the heart rate data as the ground truth. Our results exhibit that the proposed method is more accurate (71{\%}) as compared to the baseline method (54{\%}). This demonstrates that physiologically-assessed stress levels can be estimated based on the implicit features that are gathered from the smartphone logs.",
keywords = "LF/HF, mental health, smartphone log, stress",
author = "Naoki Yamamoto and Keiichi Ochiai and Akiya Inagaki and Yusuke Fukazawa and Masatoshi Kimoto and Kazuki Kiriu and Kouhei Kaminishi and Jun Ota and Tsukasa Okimura and Yuuri Terasawa and Takaki Maeda",
year = "2019",
month = "2",
day = "26",
doi = "10.23919/ICMU.2018.8653590",
language = "English",
series = "2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018",

}

TY - GEN

T1 - Physiological Stress Level Estimation Based on Smartphone Logs

AU - Yamamoto, Naoki

AU - Ochiai, Keiichi

AU - Inagaki, Akiya

AU - Fukazawa, Yusuke

AU - Kimoto, Masatoshi

AU - Kiriu, Kazuki

AU - Kaminishi, Kouhei

AU - Ota, Jun

AU - Okimura, Tsukasa

AU - Terasawa, Yuuri

AU - Maeda, Takaki

PY - 2019/2/26

Y1 - 2019/2/26

N2 - Recently, inferring the state of people's mental health via passive mobile sensing has attracted significant attention. Previous studies have used the self-assessed stress levels as the ground truth; however, these are subjective measures. In this study, we use a physiologically-assessed stress metric to minimize the effect of participant subjectivity and further estimate it using behavioral features based on the smartphone usage logs. We initially requested the study participants (39 participants) to attach heart rate sensors for 8 hours per day and simultaneously collected continuous heart rate data and smartphone logs for 42 days. Further, we divided the participants into four types via clustering using the behavioral features derived from their smartphone sensor logs and trained each model via supervised learning using the heart rate data as the ground truth. Our results exhibit that the proposed method is more accurate (71%) as compared to the baseline method (54%). This demonstrates that physiologically-assessed stress levels can be estimated based on the implicit features that are gathered from the smartphone logs.

AB - Recently, inferring the state of people's mental health via passive mobile sensing has attracted significant attention. Previous studies have used the self-assessed stress levels as the ground truth; however, these are subjective measures. In this study, we use a physiologically-assessed stress metric to minimize the effect of participant subjectivity and further estimate it using behavioral features based on the smartphone usage logs. We initially requested the study participants (39 participants) to attach heart rate sensors for 8 hours per day and simultaneously collected continuous heart rate data and smartphone logs for 42 days. Further, we divided the participants into four types via clustering using the behavioral features derived from their smartphone sensor logs and trained each model via supervised learning using the heart rate data as the ground truth. Our results exhibit that the proposed method is more accurate (71%) as compared to the baseline method (54%). This demonstrates that physiologically-assessed stress levels can be estimated based on the implicit features that are gathered from the smartphone logs.

KW - LF/HF

KW - mental health

KW - smartphone log

KW - stress

UR - http://www.scopus.com/inward/record.url?scp=85058272228&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85058272228&partnerID=8YFLogxK

U2 - 10.23919/ICMU.2018.8653590

DO - 10.23919/ICMU.2018.8653590

M3 - Conference contribution

AN - SCOPUS:85058272228

T3 - 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018

BT - 2018 11th International Conference on Mobile Computing and Ubiquitous Network, ICMU 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -