Predicting Pulmonary Function from Phone Sensors

Qian Cheng, Joshua Juen, Shashi Bellam, Nicholas Fulara, Deanna Close, Jonathan C. Silverstein, Bruce Schatz

Research output: Contribution to journalArticle

Abstract

Introduction: Smartphones are ubiquitous, but it is unknown what physiological functions can be monitored at clinical quality. Pulmonary function is a standard measure of health status for cardiopulmonary patients. We have shown phone sensors can accurately measure walking patterns. Here we show that improved classification models can accurately predict pulmonary function, with sole inputs being motion sensors from carried phones. Subjects and Methods: Twenty-five cardiopulmonary patients performed 6-minute walk tests in pulmonary rehabilitation at a regional hospital. They carried smartphones running custom software recording phone motion. Each patient's pulmonary function was measured by spirometry. A universal model, based on support vector machine, then computed the category of function with input from signal processing features and patient demographic features. Results: All but a few of every 10-second interval for every patient was correctly predicted. The trained model perfectly computed the GOLD (Global Initiative for Chronic Obstructive Lung Disease) level 1/2/3, which is a standard classification of pulmonary function. Each level was determined to have a characteristic motion, which could be recognized from the sensor features. In addition, longitudinal changes were detected for 10 patients with multiple walk tests, except for cases with clinical instability. Conclusions: These results are encouraging toward clinical validation of passive monitors running continuously in the background, for patients in homes during daily activities. Initial testing indicates the same high accuracy as with active monitors, for patients in hospitals during walk tests. We expect patients can simply carry their phones during everyday living, while models support automatic prediction of pulmonary function for health monitoring.

Original languageEnglish (US)
Pages (from-to)913-919
Number of pages7
JournalTelemedicine and e-Health
Volume23
Issue number11
DOIs
StatePublished - Nov 2017

Fingerprint

Lung
Running
Spirometry
Chronic Obstructive Pulmonary Disease
Health Status
Walking
Software
Rehabilitation
Demography
Health
Walk Test
Smartphone

Keywords

  • Chronic disease assessment
  • Health monitoring
  • Machine learning
  • Mobile phones
  • Predictive modeling
  • Pulmonary function
  • Telemedicine

ASJC Scopus subject areas

  • Health Informatics
  • Health Information Management

Cite this

Cheng, Q., Juen, J., Bellam, S., Fulara, N., Close, D., Silverstein, J. C., & Schatz, B. (2017). Predicting Pulmonary Function from Phone Sensors. Telemedicine and e-Health, 23(11), 913-919. https://doi.org/10.1089/tmj.2017.0008

Predicting Pulmonary Function from Phone Sensors. / Cheng, Qian; Juen, Joshua; Bellam, Shashi; Fulara, Nicholas; Close, Deanna; Silverstein, Jonathan C.; Schatz, Bruce.

In: Telemedicine and e-Health, Vol. 23, No. 11, 11.2017, p. 913-919.

Research output: Contribution to journalArticle

Cheng, Q, Juen, J, Bellam, S, Fulara, N, Close, D, Silverstein, JC & Schatz, B 2017, 'Predicting Pulmonary Function from Phone Sensors', Telemedicine and e-Health, vol. 23, no. 11, pp. 913-919. https://doi.org/10.1089/tmj.2017.0008
Cheng Q, Juen J, Bellam S, Fulara N, Close D, Silverstein JC et al. Predicting Pulmonary Function from Phone Sensors. Telemedicine and e-Health. 2017 Nov;23(11):913-919. https://doi.org/10.1089/tmj.2017.0008
Cheng, Qian ; Juen, Joshua ; Bellam, Shashi ; Fulara, Nicholas ; Close, Deanna ; Silverstein, Jonathan C. ; Schatz, Bruce. / Predicting Pulmonary Function from Phone Sensors. In: Telemedicine and e-Health. 2017 ; Vol. 23, No. 11. pp. 913-919.
@article{abdd73057f904048ac23780fdad1d315,
title = "Predicting Pulmonary Function from Phone Sensors",
abstract = "Introduction: Smartphones are ubiquitous, but it is unknown what physiological functions can be monitored at clinical quality. Pulmonary function is a standard measure of health status for cardiopulmonary patients. We have shown phone sensors can accurately measure walking patterns. Here we show that improved classification models can accurately predict pulmonary function, with sole inputs being motion sensors from carried phones. Subjects and Methods: Twenty-five cardiopulmonary patients performed 6-minute walk tests in pulmonary rehabilitation at a regional hospital. They carried smartphones running custom software recording phone motion. Each patient's pulmonary function was measured by spirometry. A universal model, based on support vector machine, then computed the category of function with input from signal processing features and patient demographic features. Results: All but a few of every 10-second interval for every patient was correctly predicted. The trained model perfectly computed the GOLD (Global Initiative for Chronic Obstructive Lung Disease) level 1/2/3, which is a standard classification of pulmonary function. Each level was determined to have a characteristic motion, which could be recognized from the sensor features. In addition, longitudinal changes were detected for 10 patients with multiple walk tests, except for cases with clinical instability. Conclusions: These results are encouraging toward clinical validation of passive monitors running continuously in the background, for patients in homes during daily activities. Initial testing indicates the same high accuracy as with active monitors, for patients in hospitals during walk tests. We expect patients can simply carry their phones during everyday living, while models support automatic prediction of pulmonary function for health monitoring.",
keywords = "Chronic disease assessment, Health monitoring, Machine learning, Mobile phones, Predictive modeling, Pulmonary function, Telemedicine",
author = "Qian Cheng and Joshua Juen and Shashi Bellam and Nicholas Fulara and Deanna Close and Silverstein, {Jonathan C.} and Bruce Schatz",
year = "2017",
month = "11",
doi = "10.1089/tmj.2017.0008",
language = "English (US)",
volume = "23",
pages = "913--919",
journal = "Telemedicine Journal and e-Health",
issn = "1530-5627",
publisher = "Mary Ann Liebert Inc.",
number = "11",

}

TY - JOUR

T1 - Predicting Pulmonary Function from Phone Sensors

AU - Cheng, Qian

AU - Juen, Joshua

AU - Bellam, Shashi

AU - Fulara, Nicholas

AU - Close, Deanna

AU - Silverstein, Jonathan C.

AU - Schatz, Bruce

PY - 2017/11

Y1 - 2017/11

N2 - Introduction: Smartphones are ubiquitous, but it is unknown what physiological functions can be monitored at clinical quality. Pulmonary function is a standard measure of health status for cardiopulmonary patients. We have shown phone sensors can accurately measure walking patterns. Here we show that improved classification models can accurately predict pulmonary function, with sole inputs being motion sensors from carried phones. Subjects and Methods: Twenty-five cardiopulmonary patients performed 6-minute walk tests in pulmonary rehabilitation at a regional hospital. They carried smartphones running custom software recording phone motion. Each patient's pulmonary function was measured by spirometry. A universal model, based on support vector machine, then computed the category of function with input from signal processing features and patient demographic features. Results: All but a few of every 10-second interval for every patient was correctly predicted. The trained model perfectly computed the GOLD (Global Initiative for Chronic Obstructive Lung Disease) level 1/2/3, which is a standard classification of pulmonary function. Each level was determined to have a characteristic motion, which could be recognized from the sensor features. In addition, longitudinal changes were detected for 10 patients with multiple walk tests, except for cases with clinical instability. Conclusions: These results are encouraging toward clinical validation of passive monitors running continuously in the background, for patients in homes during daily activities. Initial testing indicates the same high accuracy as with active monitors, for patients in hospitals during walk tests. We expect patients can simply carry their phones during everyday living, while models support automatic prediction of pulmonary function for health monitoring.

AB - Introduction: Smartphones are ubiquitous, but it is unknown what physiological functions can be monitored at clinical quality. Pulmonary function is a standard measure of health status for cardiopulmonary patients. We have shown phone sensors can accurately measure walking patterns. Here we show that improved classification models can accurately predict pulmonary function, with sole inputs being motion sensors from carried phones. Subjects and Methods: Twenty-five cardiopulmonary patients performed 6-minute walk tests in pulmonary rehabilitation at a regional hospital. They carried smartphones running custom software recording phone motion. Each patient's pulmonary function was measured by spirometry. A universal model, based on support vector machine, then computed the category of function with input from signal processing features and patient demographic features. Results: All but a few of every 10-second interval for every patient was correctly predicted. The trained model perfectly computed the GOLD (Global Initiative for Chronic Obstructive Lung Disease) level 1/2/3, which is a standard classification of pulmonary function. Each level was determined to have a characteristic motion, which could be recognized from the sensor features. In addition, longitudinal changes were detected for 10 patients with multiple walk tests, except for cases with clinical instability. Conclusions: These results are encouraging toward clinical validation of passive monitors running continuously in the background, for patients in homes during daily activities. Initial testing indicates the same high accuracy as with active monitors, for patients in hospitals during walk tests. We expect patients can simply carry their phones during everyday living, while models support automatic prediction of pulmonary function for health monitoring.

KW - Chronic disease assessment

KW - Health monitoring

KW - Machine learning

KW - Mobile phones

KW - Predictive modeling

KW - Pulmonary function

KW - Telemedicine

UR - http://www.scopus.com/inward/record.url?scp=85033553415&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85033553415&partnerID=8YFLogxK

U2 - 10.1089/tmj.2017.0008

DO - 10.1089/tmj.2017.0008

M3 - Article

C2 - 28300524

AN - SCOPUS:85033553415

VL - 23

SP - 913

EP - 919

JO - Telemedicine Journal and e-Health

JF - Telemedicine Journal and e-Health

SN - 1530-5627

IS - 11

ER -