A novel approach to prediction of mild obstructive sleep disordered breathing in a population-based sample: the Sleep Heart Health Study

Abstract

This manuscript considers a data-mining approach for the prediction of mild obstructive sleep disordered breathing, defined as an elevated respiratory disturbance index (RDI), in 5,530 participants in a community-based study, the Sleep Heart Health Study. The prediction algorithm was built using modern ensemble learning algorithms, boosting in specific, which allowed for assessing potential high-dimensional interactions between predictor variables or classifiers. To evaluate the performance of the algorithm, the data were split into training and validation sets for varying thresholds for predicting the probability of a high RDI (≥7 events per hour in the given results). Based on a moderate classification threshold from the boosting algorithm, the estimated post-test odds of a high RDI were 2.20 times higher than the pre-test odds given a positive test, while the corresponding post-test odds were decreased by 52% given a negative test (sensitivity and specificity of 0.66 and 0.70, respectively). In rank order, the following variables had the largest impact on prediction performance: neck circumference, body mass index, age, snoring frequency, waist circumference, and snoring loudness.

Keywords: Sleep disorders; machine learning; prediction; sleep apnea; variable importance.

Figures

Figure 1

Estimated ROC curve for the boosting algorithm from the validation data for predicting an RDI ≥ 7 events per h.

Figure 2

Plots displaying the distributions of the predicted probability of disease from the boosting algorithm for subjects with an actual RDI > 7 events/h (dashed) and

Figure 3

Plots displaying the distributions of the predicted probability of disease from the boosting algorithm for subjects with an actual RDI > 9 events/h (dashed) and

Figure 4

Variable importance plot for boosting predictions for the top 15 most influential predictors based on the validation data set. Variable names (compare with Table 1) are neck = neck circumference; BMI = body mass index; age = age in years; Snore frequency = response to the question “How often do your snore?”; Waist = waist circumference; Snore loud = response to “How loud is your snoring?”; Gender = gender of participant; Minutes = minutes to fall asleep; Sit & read = response to the question “What is the chance that you would doze off or fall asleep while sitting and reading?”; MI = MD said patient had a heart attack; HTN Meds = whether or not the participant is taking anti-hypertensive medications; SBP = systolic blood pressure; In car = response to the question “What is chance that you would doze off or fall asleep while in a car while stopped for a few minutes in traffic?”; CA = MD said patient had coronary angioplasty; TST = total sleep time.