0517 Integrating Body Sensor into a Wearable Platform to Enhance the Identification of Central and Mixed Apneas

Abstract Introduction Accurate identification of apnea types is crucial for effective diagnosis and management of sleep-disordered breathing. The Belun Sleep System (BLS-100, a.k.a., Belun Ring) is an FDA-cleared home sleep apnea testing system (K222579) comprising an adjustable ring-shaped wearable, a cradle, and two deep learning-powered algorithms. The Belun Cor, a novel subxiphoid sensor equipped with accelerometry, can detect respiratory effort, respiratory rate, body position, and facilitates the detection of central events. This preliminary analysis aims to assess the performance of the integrated BLS-100 in detecting apnea events containing central components. Methods This interim analysis evaluated the performance of BLS-100 in a clinical cohort of hospitalized patients admitted for acute ischemic stroke. Eligible patients underwent in-lab polysomnography (PSG) alongside concurrent BLS-100 testing. PSG scoring adhered to the latest AASM scoring manual, with scoring technicians blinded to the BLS-100 results. The BLS-100 derived total sleep time (bTST), sleep stages (bSTAGE), apnea-hypopnea index (bAHI), and combined central and mixed apnea index (bCMAI). Results As of 12/17/2023, 25 consecutive Taiwanese patients were enrolled. Four patients were excluded due to short bTST 120 mins. The analysis was conducted on 21 patients. M:F 19:2; age 59.7; PSG TST 270 ± 61.9 mins; PSG AHI 27.0 (1.4-81.9) with 3 normal, 3 mild, 7 moderate, and 8 severe OSA cases. The mean PSG central apnea index (PSG-CAI) was 4.8 (0.0-34.0), with 5 patients having PSG-CAI≥5. The mean PSG central and mixed apnea index (PSG-CMAI) was 8.4 (0.0-47.3). Pearson correlation coefficients between PSG-CAI and bCMAI, as well as PSG-CMAI and bCMAI, were 0.939 (P 0.001) and 0.982 (P 0.001), respectively. Using bCMAI≥5 to predict PSG-CAI≥5, the accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and Cohen’s Kappa were 0.81, 1.00, 0.75, 0.56, 1.00, and 0.59, respectively. Similarly, using bCMAI≥5 to predict PSG-CMAI≥5, the corresponding values were 0.86, 0.88, 0.85, 0.78, 0.92, and 0.70, respectively. Conclusion Early findings indicate that the BLS-100 with Belun Cor shows promising performance in identifying apnea events that include central components. An elevated bCMAI serves as a valuable indicator for clinicians, signaling the presence of central or mixed apneas. Support (if any) Case Western Reserve University-Taipei Medical University Pilot Award