Machine learning for nocturnal mass diagnosis of atrial fibrillation in a population at risk of sleep-disordered breathing

Abstract

Objective: In this research, we introduce a new methodology for atrial fibrillation (AF) diagnosis during sleep in a large population sample at risk of sleep-disordered breathing. Approach: The approach leverages digital biomarkers and recent advances in machine learning (ML) for mass AF diagnosis from overnight-hours of single-channel electrocardiogram (ECG) recording. Four databases, totaling n = 3088 patients and p = 26 913 h of continuous single-channel electrocardiogram raw data were used. Three of the databases (n = 125, p = 2513) were used for training a ML model in recognizing AF events from beat-to-beat time series. Visit 1 of the sleep heart health study database (SHHS1, n = 2963, p = 24 400) was used as the test set to evaluate the feasibility of identifying prominent AF from polysomnographic recordings. By combining AF diagnosis history and a cardiologist’s visual inspection of individuals suspected of having AF (n = 118), a total of 70 patients were diagnosed with prominent AF in SHHS1. Main results: Model prediction on SHHS1 showed an overall S e = 0.97 , S p = 0.99 , NPV = 0.99 and PPV = 0.67 in classifying individuals with or without prominent AF. PPV was non-inferior (p = 0.03) for individuals with an apnea-hypopnea index (AHI) ≥15 versus AHI < 15 . Over 22% of correctly identified prominent AF rhythm cases were not previously documented as AF in SHHS1. Significance: Individuals with prominent AF can be automatically diagnosed from an overnight single-channel ECG recording, with an accuracy unaffected by the presence of moderate-to-severe obstructive sleep apnea. This approach enables identifying a large proportion of AF individuals that were otherwise missed by regular care.

From sleep medicine to medicine during sleep–a clinical perspective

“…Holter ECG, single-lead ECG in PSG and in cases of implanted pacemakers. In a recent research of ours (Chocron et al 2020), we demonstrated that over 22% of individuals with undiagnosed atrial fibrillation could be identified by opportunistic data-driven nocturnal screening of the ECG traces recorded in regular PSG studies. Research has also shown that atrial fibrillation events may be more frequent during sleep than daytime (Yamashita et al 1997), but this may not be the case for other cardiac abnormalities (Portaluppi and Hermida 2007).…”

From sleep medicine to medicine during sleep: a new paradigm

“…Transient desaturations (even to 70% or less) are commonly observed in infants with bronchiolitis after discharge, but their clinical significance remains unclear 28 In clinical practice, there are a few instances where nocturnal ECG is recorded, e.g., Holter ECG, single-lead ECG in polysomnography (PSG) and in cases of implanted pacemakers. In a recent research of ours 33 , we demonstrated that over 22% of individuals with undiagnosed atrial fibrillation could be identified by opportunistic data-driven nocturnal screening of the ECG traces recorded in regular PSG studies. Research has also shown that atrial fibrillation events may be more frequent during sleep than daytime 34 , but this may not be case for other cardiac abnormalities 35 .…”

“…Broadening our knowledge of pathophysiological patterns during sleep may enable harnessing of machine learning algorithms to support screening strategies in selected populations. For example, opportunistic nocturnal screening for hypertension, atrial fibrillation 33 or COPD 27 , may be performed in high risk patient populations.…”

SleepPPG-Net: a deep learning algorithm for robust sleep staging from continuous photoplethysmography

“…For SHHS, the ECG peaks were detected using epltd0 [46] a state-of-the-art ECG peak detection algorithm. PRV and HRV measures were extracted using the Python HRV features implemented in [47]. This library calculates 21 HRV measures per set of IBIs.…”