Using oximetry dynamics to screen for sleep disordered breathing at varying thresholds of severity

Sleep disordered breathing (SDB) is highly prevalent in children and causes daytime sleepiness, growth failure and developmental delay. Polysomnography (PSG), the gold standard to diagnose SDB, provides an estimate of severity, called the Apnea Hypoapnea Index (AHI). PSG is costly and resource intensive; therefore we propose using the Phone Oximeter, a pulse oximeter integrated into a phone that measures blood oxygen saturation (SpO₂), as an at-home screening tool. In clinical practice, an AHI of 2-5 indicates mild SDB, but an AHI ≥ 5 usually prompts SDB treatment. Thus, we studied the performance of the Phone Oximeter as an SDB screening tool at varying thresholds of AHI. We analyzed the SpO₂ of 146 children, recorded by the Phone Oximeter, alongside conventional PSG. Time-frequency characterization of SpO₂ dynamics resulted in identification of 77% and 86% of children with an AHI ≥ 2 and AHI ≥ 5, respectively, using a multiple logistic regression model.