Predicting Depression and Anxiety Mood by Wrist-Worn Sleep Sensor

Shuichi Fukuda, Yuki Matsuda, Yutaka Arakawa, Keiichi Yasumoto, Yuri Tani: Predicting Depression and Anxiety Mood by Wrist-Worn Sleep Sensor. In: WristSense 2020: 6th Workshop on Sensing Systems and Applications using Wrist Worn Smart Devices (WristSense 2020), 2020.

Abstract

In recent years, researches on recognizing daily behavior and psychological
/ physiological states has been actively conducted to change the behavior
of workers working in companies.
In this paper, we analyzed Occupational Health questionnaire named DAMS on
waking-up time and daily sleep data that are acquired from wearable devices
in 2--3 weeks experiment of 60 office workers working at five general
companies.
By using a machine learning method, our binary Balanced Random Forest model
predicts depression, positive, and anxiety moods in two levels, high and
low.
As a result of Leave One Person Out cross validation, it was confirmed that
our model estimated with the F1 values of depression mood: 0.776, positive
mood: 0.610, anxiety mood: 0.756.
Moreover, we evaluated the variance of the three estimations among subjects
by the box chart. It was confirmed that there is variance in estimation
accuracy for each subject.

BibTeX (Download)

@inproceedings{Fuku2003:Predicting,
title = {Predicting Depression and Anxiety Mood by Wrist-Worn Sleep Sensor},
author = {Shuichi Fukuda and Yuki Matsuda and Yutaka Arakawa and Keiichi Yasumoto and Yuri Tani},
url = {https://arakawa-lab.com/wp-content/uploads/2022/08/1570616568-stamped-e.pdf
https://ieeexplore.ieee.org/document/9156176},
year  = {2020},
date = {2020-03-23},
urldate = {2020-03-23},
booktitle = {WristSense 2020: 6th Workshop on Sensing Systems and Applications using Wrist Worn Smart Devices (WristSense 2020)},
abstract = {In recent years, researches on recognizing daily behavior and psychological 
/ physiological states has been actively conducted to change the behavior 
of workers working in companies. 
In this paper, we analyzed Occupational Health questionnaire named DAMS on 
waking-up time and daily sleep data that are acquired from wearable devices 
in 2--3 weeks experiment of 60 office workers working at five general 
companies. 
By using a machine learning method, our binary Balanced Random Forest model 
predicts depression, positive, and anxiety moods in two levels, high and 
low. 
As a result of Leave One Person Out cross validation, it was confirmed that 
our model estimated with the F1 values of depression mood: 0.776, positive 
mood: 0.610, anxiety mood: 0.756. 
Moreover, we evaluated the variance of the three estimations among subjects 
by the box chart. It was confirmed that there is variance in estimation 
accuracy for each subject.},
keywords = {stress sensing, wearable computing, Wrist Sensor; Life log; Machine learning; Wearable computing; Occupational  Health},
pubstate = {published},
tppubtype = {inproceedings}
}