475 Automatic detection of self-similarity and prediction of CPAP failure

Abstract

Abstract Introduction Sleep disordered breathing is a significant risk factor for cardiometabolic and neurodegenerative diseases. Tolerance and efficacy of continuous positive airway pressure (CPAP), the primary form of therapy for sleep apnea, is often poor. High loop gain (HLG) is a driving mechanism of central sleep apnea or periodic breathing. The current study aimed to develop a computational approach to detect HLG based on self-similarity in respiratory oscillations during sleep solely using breathing patterns, measured via Respiratory Inductance Plethysmography (RIP). To quantify the potential utility of the developed similarity metric, the presented algorithm was used to predict acute CPAP failure. Methods We developed an algorithm for detecting apneas as periods with reduced breathing effort, manifested in the RIP signal as low signal amplitude. Our algorithm calculates self-similarity in breathing patterns between consecutive periods of apnea or hypopnea. Working under the assumption that high loop gain induces self-similar respiratory oscillations and increases the risk of failure during CPAP, the full night similarity, computed during diagnostic non-CPAP polysomnography (PSG), was used to predict failure of CPAP, which we defined as titration central apnea index (CAI)>10. Central apnea labels are obtained both from manual scoring by sleep technologists, and from an automated algorithm developed for this study. The Massachusetts General Hospital (MGH) sleep database was used, including 2466 PSG pairs of diagnostic and CPAP titration PSG recordings. Results Diagnostic CAI based on technologist labels predicted failure of CPAP with an AUC of 0.82 ±0.03. Based on automatically generated labels, the combination of full night similarity and automatically generated CAI resulted in an AUC of 0.85 ±0.02. A subanalysis was performed on a population with technologist labeled diagnostic CAI>5. Full night similarity predicted failure with an AUC of 0.57 ±0.07 for manual and 0.65 ±0.06 for automated labels. Conclusion This study showed that central apnea labels can be derived in an automated way. The proposed self-similarity feature, as a surrogate estimate of expressed respiratory high loop gain and computed from easily accessible effort signals, can detect periodic breathing regardless of admixed obstructive features such as flow-limitation, and can aid prediction of CPAP failure or success. Support (if any):