0448 Scaled-Up Sleep Apnea Endotyping Using Polysomnography for Clinical Use

Abstract Introduction Sleep apnea is caused by several key endophenotypic traits namely pharyngeal collapsibility, poor muscle compensation, ventilatory instability (high loop gain), and arousability from sleep (low arousal threshold). Already, measures of these traits have shown promise for predicting outcomes of therapies (oral appliances, surgery, hypoglossal nerve stimulation, CPAP, or pharmaceuticals) and thus may be an integral part of future precision sleep medicine with treatments administered based on underlying pathophysiology. However, currently, the novel methods developed for endotyping from polysomnography are computationally expensive and can only be performed by the original authors or their collaborators due to the need for technological expertise. Here we present a cloud-based method for endotyping sleep apnea from polysomnography for use in the clinical arena. Methods For cloud-based use, we optimized the Phenotype Using Polysomnography (‘PUP’) method of Sands et.al. (2015-2018) by performing the following: Code was translated from MATLAB to Python; efficient methods were developed to detect breaths, calculate normalized ventilation (moving time-average), and model ventilatory drive (intended ventilation). The new implementation (‘PUP.py’) was validated by comparing the measured traits against the original values. Results 38 manually scored clinical polysomnographic studies were endophenotyped using the two implementations. Results of the new implementation (‘PUP.py’) were strongly correlated with the original (p