Key result
ECG-based XGBoost model detects sleep respiratory events with ~88% accuracy.
Why the study?
Calculating the apnea hypopnea index to diagnose sleep apnea hypopnea syndrome severity requires accurate identification of various sleep respiratory events.
Does an automated machine learning algorithm using electrocardiogram and cardiopulmonary signals accurately detect and classify sleep respiratory events compared to manual polysomnography annotation?
Does an automated machine learning algorithm using electrocardiogram and cardiopulmonary signals accurately detect and classify sleep respiratory events compared to manual polysomnography annotation?
Effect estimate: Accuracy 0.877
An automated machine learning algorithm fusion utilizing ECG and cardiopulmonary signals achieves high accuracy in identifying and sub-classifying sleep apnea events, providing a theoretical foundation for robust out-of-hospital sleep monitoring.
ECG-based XGBoost model may aid automated sleep event detection; leaves open prospective validation before clinical adoption.
Sleep apnea hypopnea syndrome (SAHS) is a common sleep disorder with a high prevalence. The apnea hypopnea index (AHI) is an important indicator used to diagnose the severity of SAHS disorders. The calculation of the AHI is based on the accurate identification of various types of sleep respiratory events. In this paper, we proposed an automatic detection algorithm for respiratory events during sleep. In addition to the accurate recognition of normal breathing, hypopnea and apnea events using heart rate variability (HRV), entropy and other manual features, we also presented a fusion of ribcage and abdomen movement data combined with the long short-term memory (LSTM) framework to achieve the distinction between obstructive and central apnea events. While only using electrocardiogram (ECG) features, the accuracy, precision, sensitivity, and F1 score of the XGBoost model are 0.877, 0.877, 0.876, and 0.876, respectively, demonstrating that it performs better than other models. Moreover, the accuracy, sensitivity, and F1 score of the LSTM model for detecting obstructive and central apnea events were 0.866, 0.867, and 0.866, respectively. The research results of this paper can be used for the automatic recognition of sleep respiratory events as well as AHI calculation of polysomnography (PSG), which provide a theoretical basis and algorithm references for out-of-hospital sleep monitoring.
No takes yet. Share an insight, caveat, or question.
Yan et al. (2023) studied Sleep apnea hypopnea syndrome (SAHS) (n=5,829). Machine learning automatic detection algorithm (XGBoost and LSTM) using ECG and airflow signals vs. Manual annotation by sleep specialists (polysomnography gold standard) and other standard machine learning models was evaluated on Detection of sleep respiratory events (normal breathing, hypopnea, apnea) using only ECG features with the XGBoost model (Accuracy 0.877). An automated XGBoost machine learning model utilizing only electrocardiogram features achieved an accuracy of 0.877, precision of 0.877, and sensitivity of 0.876 for detecting sleep respiratory events.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: