Design and evaluation of a non-contact bed-mounted sensing device for automated in-home detection of obstructive sleep apnea

A pilot study

Clara Mosquera-Lopez, Joseph Leitschuh, John Condon, Chad Hagen, Uma Rajhbeharrysingh, Cody Hanks, Peter Jacobs

Research output: Contribution to journalArticle

Abstract

We conducted a pilot study to evaluate the accuracy of a custom built non-contact pressure-sensitive device in diagnosing obstructive sleep apnea (OSA) severity as an alternative to in-laboratory polysomnography (PSG) and a Type 3 in-home sleep apnea test (HSAT). Fourteen patients completed PSG sleep studies for one night with simultaneous recording from our load-cell-based sensing device in the bed. Subjects subsequently installed pressure sensors in their bed at home and recorded signals for up to four nights. Machine learning models were optimized to classify sleep apnea severity using a standardized American Academy of Sleep Medicine (AASM) scoring of the gold standard studies as reference. On a per-night basis, our model reached a correct OSA detection rate of 82.9% (sensitivity = 88.9%, specificity = 76.5%), and OSA severity classification accuracy of 74.3% (61.5% and 81.8% correctly classified in-clinic and in-home tests, respectively). There was no difference in Apnea Hypopnea Index (AHI) estimation when subjects wore HSAT sensors versus load cells (LCs) only (p-value = 0.62). Our in-home diagnostic system provides an unobtrusive method for detecting OSA with high sensitivity and may potentially be used for long-term monitoring of breathing during sleep. Further research is needed to address the lower specificity resulting from using the highest AHI from repeated samples.

Original languageEnglish (US)
Article number90
JournalBiosensors
Volume9
Issue number3
DOIs
StatePublished - Sep 1 2019

Fingerprint

Obstructive Sleep Apnea
Sleep Apnea Syndromes
Equipment and Supplies
Sleep
Polysomnography
Apnea
Pressure
Respiration
Medicine
Sensitivity and Specificity
Research
Pressure sensors
Learning systems
Wear of materials
Monitoring
Sensors

Keywords

  • Apnea-Hypopnea index
  • Automated obstructive sleep apnea diagnosis
  • Contact-less load cell sensor
  • Long-term breathing monitoring
  • Unobtrusive obstructive sleep apnea diagnosis

ASJC Scopus subject areas

  • Clinical Biochemistry

Cite this

Design and evaluation of a non-contact bed-mounted sensing device for automated in-home detection of obstructive sleep apnea : A pilot study. / Mosquera-Lopez, Clara; Leitschuh, Joseph; Condon, John; Hagen, Chad; Rajhbeharrysingh, Uma; Hanks, Cody; Jacobs, Peter.

In: Biosensors, Vol. 9, No. 3, 90, 01.09.2019.

Research output: Contribution to journalArticle

Mosquera-Lopez, Clara ; Leitschuh, Joseph ; Condon, John ; Hagen, Chad ; Rajhbeharrysingh, Uma ; Hanks, Cody ; Jacobs, Peter. / Design and evaluation of a non-contact bed-mounted sensing device for automated in-home detection of obstructive sleep apnea : A pilot study. In: Biosensors. 2019 ; Vol. 9, No. 3.
@article{b54bd58d19444728a54a2baaa0fc32f9,
title = "Design and evaluation of a non-contact bed-mounted sensing device for automated in-home detection of obstructive sleep apnea: A pilot study",
abstract = "We conducted a pilot study to evaluate the accuracy of a custom built non-contact pressure-sensitive device in diagnosing obstructive sleep apnea (OSA) severity as an alternative to in-laboratory polysomnography (PSG) and a Type 3 in-home sleep apnea test (HSAT). Fourteen patients completed PSG sleep studies for one night with simultaneous recording from our load-cell-based sensing device in the bed. Subjects subsequently installed pressure sensors in their bed at home and recorded signals for up to four nights. Machine learning models were optimized to classify sleep apnea severity using a standardized American Academy of Sleep Medicine (AASM) scoring of the gold standard studies as reference. On a per-night basis, our model reached a correct OSA detection rate of 82.9{\%} (sensitivity = 88.9{\%}, specificity = 76.5{\%}), and OSA severity classification accuracy of 74.3{\%} (61.5{\%} and 81.8{\%} correctly classified in-clinic and in-home tests, respectively). There was no difference in Apnea Hypopnea Index (AHI) estimation when subjects wore HSAT sensors versus load cells (LCs) only (p-value = 0.62). Our in-home diagnostic system provides an unobtrusive method for detecting OSA with high sensitivity and may potentially be used for long-term monitoring of breathing during sleep. Further research is needed to address the lower specificity resulting from using the highest AHI from repeated samples.",
keywords = "Apnea-Hypopnea index, Automated obstructive sleep apnea diagnosis, Contact-less load cell sensor, Long-term breathing monitoring, Unobtrusive obstructive sleep apnea diagnosis",
author = "Clara Mosquera-Lopez and Joseph Leitschuh and John Condon and Chad Hagen and Uma Rajhbeharrysingh and Cody Hanks and Peter Jacobs",
year = "2019",
month = "9",
day = "1",
doi = "10.3390/bios9030090",
language = "English (US)",
volume = "9",
journal = "Biosensors",
issn = "0956-5663",
publisher = "Elsevier Limited",
number = "3",

}

TY - JOUR

T1 - Design and evaluation of a non-contact bed-mounted sensing device for automated in-home detection of obstructive sleep apnea

T2 - A pilot study

AU - Mosquera-Lopez, Clara

AU - Leitschuh, Joseph

AU - Condon, John

AU - Hagen, Chad

AU - Rajhbeharrysingh, Uma

AU - Hanks, Cody

AU - Jacobs, Peter

PY - 2019/9/1

Y1 - 2019/9/1

N2 - We conducted a pilot study to evaluate the accuracy of a custom built non-contact pressure-sensitive device in diagnosing obstructive sleep apnea (OSA) severity as an alternative to in-laboratory polysomnography (PSG) and a Type 3 in-home sleep apnea test (HSAT). Fourteen patients completed PSG sleep studies for one night with simultaneous recording from our load-cell-based sensing device in the bed. Subjects subsequently installed pressure sensors in their bed at home and recorded signals for up to four nights. Machine learning models were optimized to classify sleep apnea severity using a standardized American Academy of Sleep Medicine (AASM) scoring of the gold standard studies as reference. On a per-night basis, our model reached a correct OSA detection rate of 82.9% (sensitivity = 88.9%, specificity = 76.5%), and OSA severity classification accuracy of 74.3% (61.5% and 81.8% correctly classified in-clinic and in-home tests, respectively). There was no difference in Apnea Hypopnea Index (AHI) estimation when subjects wore HSAT sensors versus load cells (LCs) only (p-value = 0.62). Our in-home diagnostic system provides an unobtrusive method for detecting OSA with high sensitivity and may potentially be used for long-term monitoring of breathing during sleep. Further research is needed to address the lower specificity resulting from using the highest AHI from repeated samples.

AB - We conducted a pilot study to evaluate the accuracy of a custom built non-contact pressure-sensitive device in diagnosing obstructive sleep apnea (OSA) severity as an alternative to in-laboratory polysomnography (PSG) and a Type 3 in-home sleep apnea test (HSAT). Fourteen patients completed PSG sleep studies for one night with simultaneous recording from our load-cell-based sensing device in the bed. Subjects subsequently installed pressure sensors in their bed at home and recorded signals for up to four nights. Machine learning models were optimized to classify sleep apnea severity using a standardized American Academy of Sleep Medicine (AASM) scoring of the gold standard studies as reference. On a per-night basis, our model reached a correct OSA detection rate of 82.9% (sensitivity = 88.9%, specificity = 76.5%), and OSA severity classification accuracy of 74.3% (61.5% and 81.8% correctly classified in-clinic and in-home tests, respectively). There was no difference in Apnea Hypopnea Index (AHI) estimation when subjects wore HSAT sensors versus load cells (LCs) only (p-value = 0.62). Our in-home diagnostic system provides an unobtrusive method for detecting OSA with high sensitivity and may potentially be used for long-term monitoring of breathing during sleep. Further research is needed to address the lower specificity resulting from using the highest AHI from repeated samples.

KW - Apnea-Hypopnea index

KW - Automated obstructive sleep apnea diagnosis

KW - Contact-less load cell sensor

KW - Long-term breathing monitoring

KW - Unobtrusive obstructive sleep apnea diagnosis

UR - http://www.scopus.com/inward/record.url?scp=85070446448&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85070446448&partnerID=8YFLogxK

U2 - 10.3390/bios9030090

DO - 10.3390/bios9030090

M3 - Article

VL - 9

JO - Biosensors

JF - Biosensors

SN - 0956-5663

IS - 3

M1 - 90

ER -