Novel method combining multiscale attention entropy of overnight blood oxygen level and machine learning for easy sleep apnea screening

Abstract

Objective Sleep apnea is a common sleep disorder affecting a significant portion of the population, but many apnea patients remain undiagnosed because existing clinical tests are invasive and expensive. This study aimed to develop a method for easy sleep apnea screening. Methods Three supervised machine learning algorithms, including logistic regression, support vector machine, and light gradient boosting machine, were applied to develop apnea screening models at two apnea–hypopnea index cutoff thresholds: [Formula: see text] 5 and [Formula: see text] 30 events/hours. The SpO2 recordings of the Sleep Heart Health Study database ( N = 5786) were used for model training, validation, and test. Multiscale entropy analysis was performed to derive a set of multiscale attention entropy features from the SpO2 recordings. Demographic features including age, sex, body mass index, and blood pressure were also used. The dependency among the multiscale attention entropy features were handled with the independent component analysis. Results For cutoff [Formula: see text] 5/hours, logistic regression model achieved the highest Matthew’s correlation coefficient (0.402) and area under the curve (0.747), and reasonably good sensitivity (75.38%), specificity (74.02%), and positive predictive value (92.94%). For cutoff [Formula: see text] 30/hours, support vector machine model achieved the highest Matthew’s correlation coefficient (0.545) and area under the curve (0.823), and good sensitivity (82.00%), specificity (82.69%), and negative predictive value (95.53%). Conclusions Our models achieved better performance than existing methods and have the potential to be integrated with home-use pulse oximeters.

Developing Probabilistic Ensemble Machine Learning Models for Home-Based Sleep Apnea Screening using Overnight SpO2 Data at Varying Data Granularity

“…In another study (N=6,875) [27], three models were developed at all three AHI cutoffs, achieving reasonably high AUCs (0.80-0.82), good sensitivity (0.74-0.80), and slightly low specificity (0.62-0.75). Similarly, another study (N=5,786) built large models at cutoffs of ≥5 and ≥30 achieved reasonably high AUC (0.75-0.82), good sensitivity (0.75-0.82), and good specificity (0.74-0.83) [9].…”

“…For each AHI cutoff point, we constructed a probabilistic ensemble model comprising three base classifiers: support vector machine (SVM), logistic regression (LR), and light gradient boosting machine (LGBM). These machine learning algorithms have exhibited promising performance in sleep apnea screening in previous studies [9]. SVM operates by mapping a low-dimensional input feature space to a higher-dimensional domain, where it identifies a hyperplane positioned as far as possible from the marginal samples of each class (i.e., the support vectors).…”

“…Numerous studies have leveraged a subset of signals derived from PSG, such as electroencephalogram (EEG), electrocardiogram (ECG), airflow, and blood oxygen saturation levels (SpO2), for automated sleep apnea screening. Performance outcomes vary based on signal modality and computational models utilized [4][5][6][7][8][9][10][11][12][13]. Yet, most sensing modalities, including EEG, ECG, and airflow, are not readily available for home use, hindering widespread adoption for at-home apnea screening.…”

Developing and Validating Ensemble Classifiers for At-Home Sleep Apnea Screening

“…We built two ensemble classifiers: f ecl f 1 uses a weight vector of [1, 1, 1, 1, 1], while f ecl f 2 uses a weight vector of [2, 1, 2, 1, 1]. We assigned heavier weights to the LR and SVM classifiers in the latter case because those two classifiers yielded better performance in our previous study [8].…”