Efficient Screening in Obstructive Sleep Apnea Using Sequential Machine Learning Models, Questionnaires, and Pulse Oximetry Signals: Mixed Methods Study
2024

Efficient Screening for Sleep Apnea Using Machine Learning

Sample size: 9663 publication 10 minutes Evidence: moderate

Author Information

Author(s): Amy Schwartz, Jean-Marie Aerts, Lei Guo, Nai-Yu Kuo, Hsin-Jung Tsai, Shih-Jen Tsai, Albert C Yang

Primary Institution: National Yang Ming Chiao Tung University

Hypothesis

This study aims to develop 2 sequential machine learning models to efficiently screen and differentiate obstructive sleep apnea (OSA).

Conclusion

The study developed two machine learning models that effectively screen for sleep apnea, demonstrating good predictive ability.

Supporting Evidence

  • The Model-Questionnaire achieved an F1-score of 0.86.
  • The Model-Saturation reached an F1-score of 0.85.
  • Both models demonstrated adequate at-home screening capabilities for sleep disorders.

Takeaway

Researchers created two computer models to help doctors quickly find out if someone has sleep apnea, a condition that makes it hard to breathe while sleeping.

Methodology

The study used two datasets, one from the Sleep Heart Health Study and another from Taipei Veterans General Hospital, to train and test machine learning models based on questionnaires and blood oxygen levels.

Potential Biases

Potential biases may arise from using datasets from different geographic areas, which could affect model performance.

Limitations

The study lacked a healthy control group and may not generalize well to younger populations or those outside the studied demographics.

Participant Demographics

The study included a diverse population with varying ages, genders, and health conditions, primarily focusing on adults over 18.

Digital Object Identifier (DOI)

10.2196/51615

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication