Automatic Prediction of Health Status using Smartphone Derived Behavior Profiles

Daniel Kelly, Kevin Curran, Brian Caulfield

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Objective: Current methods of assessing the affect a patients’ health has on their daily life are extremely limited. The aim of this work is to develop a sensor based approach to health status measurement in order to objectively measure health status.

Methods: Techniques to generate human behaviour profiles, derived from smartphone accelerometer and gyroscope sensors, are proposed. Experiments, using SVM regression models, are then conducted in order to evaluate the use of the proposed behaviour profiles as a predictor of health status.

Results: Experiments were conducted on data from 171 participants, with an average of 114 hours of data per participant. Regression models were trained and tested on the 10 SF-36 self-ratings. Results showed that the 8 individual SF-36 scales and 2 component scores could be predicted with an average correlation of 0.683 and 0.698 respectively. General Health was predicted with an average correlation of 0.752.

Conclusion: Research shows that the Clinically Important Difference for SF-36 self-ratings are approximately 10 points. Health status prediction errors in this work were 11.7 points on average. While the problem has not been fully solved, this work present a hugely promising direction for health status prediction.

Significance: Using the proposed techniques, health status could be measured using unobtrusive, inexpensive and already available hardware. It could provide a means for clinicians to accurately and objectively assess the daily life benefits of treatments on an individual patient basis.
LanguageEnglish
Pages1750-1760
Number of pages11
JournalIEEE Journal of Biomedical and Health Informatics
Volume21
Issue number6
Early online date9 Jan 2017
DOIs
Publication statusPublished - Nov 2017

Fingerprint

Smartphones
Health Status
Health
Personnel rating
Smartphone
Gyroscopes
Sensors
Accelerometers
Experiments
Hardware
Research

Keywords

  • Smartphone
  • Health Status

Cite this

@article{e6771512bcea47869e1d428f833a72de,
title = "Automatic Prediction of Health Status using Smartphone Derived Behavior Profiles",
abstract = "Objective: Current methods of assessing the affect a patients’ health has on their daily life are extremely limited. The aim of this work is to develop a sensor based approach to health status measurement in order to objectively measure health status.Methods: Techniques to generate human behaviour profiles, derived from smartphone accelerometer and gyroscope sensors, are proposed. Experiments, using SVM regression models, are then conducted in order to evaluate the use of the proposed behaviour profiles as a predictor of health status. Results: Experiments were conducted on data from 171 participants, with an average of 114 hours of data per participant. Regression models were trained and tested on the 10 SF-36 self-ratings. Results showed that the 8 individual SF-36 scales and 2 component scores could be predicted with an average correlation of 0.683 and 0.698 respectively. General Health was predicted with an average correlation of 0.752. Conclusion: Research shows that the Clinically Important Difference for SF-36 self-ratings are approximately 10 points. Health status prediction errors in this work were 11.7 points on average. While the problem has not been fully solved, this work present a hugely promising direction for health status prediction. Significance: Using the proposed techniques, health status could be measured using unobtrusive, inexpensive and already available hardware. It could provide a means for clinicians to accurately and objectively assess the daily life benefits of treatments on an individual patient basis.",
keywords = "Smartphone, Health Status",
author = "Daniel Kelly and Kevin Curran and Brian Caulfield",
year = "2017",
month = "11",
doi = "10.1109/JBHI.2017.2649602",
language = "English",
volume = "21",
pages = "1750--1760",
journal = "IEEE Journal of Biomedical and Health Informatics",
issn = "2168-2194",
number = "6",

}

Automatic Prediction of Health Status using Smartphone Derived Behavior Profiles. / Kelly, Daniel; Curran, Kevin; Caulfield, Brian.

In: IEEE Journal of Biomedical and Health Informatics, Vol. 21, No. 6, 11.2017, p. 1750-1760.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Automatic Prediction of Health Status using Smartphone Derived Behavior Profiles

AU - Kelly, Daniel

AU - Curran, Kevin

AU - Caulfield, Brian

PY - 2017/11

Y1 - 2017/11

N2 - Objective: Current methods of assessing the affect a patients’ health has on their daily life are extremely limited. The aim of this work is to develop a sensor based approach to health status measurement in order to objectively measure health status.Methods: Techniques to generate human behaviour profiles, derived from smartphone accelerometer and gyroscope sensors, are proposed. Experiments, using SVM regression models, are then conducted in order to evaluate the use of the proposed behaviour profiles as a predictor of health status. Results: Experiments were conducted on data from 171 participants, with an average of 114 hours of data per participant. Regression models were trained and tested on the 10 SF-36 self-ratings. Results showed that the 8 individual SF-36 scales and 2 component scores could be predicted with an average correlation of 0.683 and 0.698 respectively. General Health was predicted with an average correlation of 0.752. Conclusion: Research shows that the Clinically Important Difference for SF-36 self-ratings are approximately 10 points. Health status prediction errors in this work were 11.7 points on average. While the problem has not been fully solved, this work present a hugely promising direction for health status prediction. Significance: Using the proposed techniques, health status could be measured using unobtrusive, inexpensive and already available hardware. It could provide a means for clinicians to accurately and objectively assess the daily life benefits of treatments on an individual patient basis.

AB - Objective: Current methods of assessing the affect a patients’ health has on their daily life are extremely limited. The aim of this work is to develop a sensor based approach to health status measurement in order to objectively measure health status.Methods: Techniques to generate human behaviour profiles, derived from smartphone accelerometer and gyroscope sensors, are proposed. Experiments, using SVM regression models, are then conducted in order to evaluate the use of the proposed behaviour profiles as a predictor of health status. Results: Experiments were conducted on data from 171 participants, with an average of 114 hours of data per participant. Regression models were trained and tested on the 10 SF-36 self-ratings. Results showed that the 8 individual SF-36 scales and 2 component scores could be predicted with an average correlation of 0.683 and 0.698 respectively. General Health was predicted with an average correlation of 0.752. Conclusion: Research shows that the Clinically Important Difference for SF-36 self-ratings are approximately 10 points. Health status prediction errors in this work were 11.7 points on average. While the problem has not been fully solved, this work present a hugely promising direction for health status prediction. Significance: Using the proposed techniques, health status could be measured using unobtrusive, inexpensive and already available hardware. It could provide a means for clinicians to accurately and objectively assess the daily life benefits of treatments on an individual patient basis.

KW - Smartphone

KW - Health Status

U2 - 10.1109/JBHI.2017.2649602

DO - 10.1109/JBHI.2017.2649602

M3 - Article

VL - 21

SP - 1750

EP - 1760

JO - IEEE Journal of Biomedical and Health Informatics

T2 - IEEE Journal of Biomedical and Health Informatics

JF - IEEE Journal of Biomedical and Health Informatics

SN - 2168-2194

IS - 6

ER -