ML approach for early detection of sleep apnea treatment abandonment: A case study

Matheus Araujo, Rahul Bhojwani, Jaideep Srivastava, Louis Kazaglis, Conrad Iber

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

1 Citation (Scopus)

Abstract

Sleep apnea is a growing problem in the country, with over 200,000 new cases being identified each year. Continuous positive airway pressure (CPAP) is the best treatment for obstructive sleep apnea (OSA), but is limited by low adherence to treatment. Fairview’s Sleep program actively tracks CPAP usage and outcomes and employs tele-health coaching to improve adherence. This labor-intensive protocol is applied to those who are failing to meet early adherence targets. However, the implementation of this is based on heuristic rules which may not be matched to actual outcomes, contacting some patients too late and others unnecessarily. Machine learning can facilitate efficient contact strategies through early and accurate identification of therapy trajectories based on patient history, including EHR data, health information, questionnaires, and daily PAP metrics. Prediction models for classification of patients regarding CPAP adherence at a clinically-important time of 6 months of regular use were built. Using data from the first 30 days of CPAP usage, and a more aggressive decision scenario from the first 13 days of usage, the proposed approach results in an improvement in prediction significantly better than the current approach used by the hospital. Further, it is shown that a hospital can utilize this precise and earlier prediction by implementing appropriate actions based on the patient’s predicted risk level.

Original languageEnglish (US)
Title of host publicationDH 2018 - Proceedings of the 2018 International Conference on Digital Health
PublisherAssociation for Computing Machinery
Pages75-79
Number of pages5
ISBN (Electronic)9781450364935
DOIs
StatePublished - Apr 23 2018
Event8th International Conference on Digital Health, DH 2018 - Lyon, France
Duration: Apr 23 2018Apr 26 2018

Publication series

NameACM International Conference Proceeding Series
Volume2018-April

Other

Other8th International Conference on Digital Health, DH 2018
CountryFrance
CityLyon
Period4/23/184/26/18

Fingerprint

Health
Learning systems
Trajectories
Personnel
Sleep

Keywords

  • CPAP adherence
  • Health informatics
  • Machine learning
  • Sleep apnea treatment

Cite this

Araujo, M., Bhojwani, R., Srivastava, J., Kazaglis, L., & Iber, C. (2018). ML approach for early detection of sleep apnea treatment abandonment: A case study. In DH 2018 - Proceedings of the 2018 International Conference on Digital Health (pp. 75-79). (ACM International Conference Proceeding Series; Vol. 2018-April). Association for Computing Machinery. https://doi.org/10.1145/3194658.3194681

ML approach for early detection of sleep apnea treatment abandonment : A case study. / Araujo, Matheus; Bhojwani, Rahul; Srivastava, Jaideep; Kazaglis, Louis; Iber, Conrad.

DH 2018 - Proceedings of the 2018 International Conference on Digital Health. Association for Computing Machinery, 2018. p. 75-79 (ACM International Conference Proceeding Series; Vol. 2018-April).

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

Araujo, M, Bhojwani, R, Srivastava, J, Kazaglis, L & Iber, C 2018, ML approach for early detection of sleep apnea treatment abandonment: A case study. in DH 2018 - Proceedings of the 2018 International Conference on Digital Health. ACM International Conference Proceeding Series, vol. 2018-April, Association for Computing Machinery, pp. 75-79, 8th International Conference on Digital Health, DH 2018, Lyon, France, 4/23/18. https://doi.org/10.1145/3194658.3194681
Araujo M, Bhojwani R, Srivastava J, Kazaglis L, Iber C. ML approach for early detection of sleep apnea treatment abandonment: A case study. In DH 2018 - Proceedings of the 2018 International Conference on Digital Health. Association for Computing Machinery. 2018. p. 75-79. (ACM International Conference Proceeding Series). https://doi.org/10.1145/3194658.3194681
Araujo, Matheus ; Bhojwani, Rahul ; Srivastava, Jaideep ; Kazaglis, Louis ; Iber, Conrad. / ML approach for early detection of sleep apnea treatment abandonment : A case study. DH 2018 - Proceedings of the 2018 International Conference on Digital Health. Association for Computing Machinery, 2018. pp. 75-79 (ACM International Conference Proceeding Series).
@inproceedings{9a607fdbed8342babdeb90a3aa499016,
title = "ML approach for early detection of sleep apnea treatment abandonment: A case study",
abstract = "Sleep apnea is a growing problem in the country, with over 200,000 new cases being identified each year. Continuous positive airway pressure (CPAP) is the best treatment for obstructive sleep apnea (OSA), but is limited by low adherence to treatment. Fairview’s Sleep program actively tracks CPAP usage and outcomes and employs tele-health coaching to improve adherence. This labor-intensive protocol is applied to those who are failing to meet early adherence targets. However, the implementation of this is based on heuristic rules which may not be matched to actual outcomes, contacting some patients too late and others unnecessarily. Machine learning can facilitate efficient contact strategies through early and accurate identification of therapy trajectories based on patient history, including EHR data, health information, questionnaires, and daily PAP metrics. Prediction models for classification of patients regarding CPAP adherence at a clinically-important time of 6 months of regular use were built. Using data from the first 30 days of CPAP usage, and a more aggressive decision scenario from the first 13 days of usage, the proposed approach results in an improvement in prediction significantly better than the current approach used by the hospital. Further, it is shown that a hospital can utilize this precise and earlier prediction by implementing appropriate actions based on the patient’s predicted risk level.",
keywords = "CPAP adherence, Health informatics, Machine learning, Sleep apnea treatment",
author = "Matheus Araujo and Rahul Bhojwani and Jaideep Srivastava and Louis Kazaglis and Conrad Iber",
year = "2018",
month = "4",
day = "23",
doi = "10.1145/3194658.3194681",
language = "English (US)",
series = "ACM International Conference Proceeding Series",
publisher = "Association for Computing Machinery",
pages = "75--79",
booktitle = "DH 2018 - Proceedings of the 2018 International Conference on Digital Health",

}

TY - GEN

T1 - ML approach for early detection of sleep apnea treatment abandonment

T2 - A case study

AU - Araujo, Matheus

AU - Bhojwani, Rahul

AU - Srivastava, Jaideep

AU - Kazaglis, Louis

AU - Iber, Conrad

PY - 2018/4/23

Y1 - 2018/4/23

N2 - Sleep apnea is a growing problem in the country, with over 200,000 new cases being identified each year. Continuous positive airway pressure (CPAP) is the best treatment for obstructive sleep apnea (OSA), but is limited by low adherence to treatment. Fairview’s Sleep program actively tracks CPAP usage and outcomes and employs tele-health coaching to improve adherence. This labor-intensive protocol is applied to those who are failing to meet early adherence targets. However, the implementation of this is based on heuristic rules which may not be matched to actual outcomes, contacting some patients too late and others unnecessarily. Machine learning can facilitate efficient contact strategies through early and accurate identification of therapy trajectories based on patient history, including EHR data, health information, questionnaires, and daily PAP metrics. Prediction models for classification of patients regarding CPAP adherence at a clinically-important time of 6 months of regular use were built. Using data from the first 30 days of CPAP usage, and a more aggressive decision scenario from the first 13 days of usage, the proposed approach results in an improvement in prediction significantly better than the current approach used by the hospital. Further, it is shown that a hospital can utilize this precise and earlier prediction by implementing appropriate actions based on the patient’s predicted risk level.

AB - Sleep apnea is a growing problem in the country, with over 200,000 new cases being identified each year. Continuous positive airway pressure (CPAP) is the best treatment for obstructive sleep apnea (OSA), but is limited by low adherence to treatment. Fairview’s Sleep program actively tracks CPAP usage and outcomes and employs tele-health coaching to improve adherence. This labor-intensive protocol is applied to those who are failing to meet early adherence targets. However, the implementation of this is based on heuristic rules which may not be matched to actual outcomes, contacting some patients too late and others unnecessarily. Machine learning can facilitate efficient contact strategies through early and accurate identification of therapy trajectories based on patient history, including EHR data, health information, questionnaires, and daily PAP metrics. Prediction models for classification of patients regarding CPAP adherence at a clinically-important time of 6 months of regular use were built. Using data from the first 30 days of CPAP usage, and a more aggressive decision scenario from the first 13 days of usage, the proposed approach results in an improvement in prediction significantly better than the current approach used by the hospital. Further, it is shown that a hospital can utilize this precise and earlier prediction by implementing appropriate actions based on the patient’s predicted risk level.

KW - CPAP adherence

KW - Health informatics

KW - Machine learning

KW - Sleep apnea treatment

UR - http://www.scopus.com/inward/record.url?scp=85047211715&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85047211715&partnerID=8YFLogxK

U2 - 10.1145/3194658.3194681

DO - 10.1145/3194658.3194681

M3 - Conference contribution

AN - SCOPUS:85047211715

T3 - ACM International Conference Proceeding Series

SP - 75

EP - 79

BT - DH 2018 - Proceedings of the 2018 International Conference on Digital Health

PB - Association for Computing Machinery

ER -