Introduction: Breathing asynchrony events during NIV include ineffective efforts, double triggering, auto-triggering, and other pre- and post-cycling asynchronies (1,2). In some situations, breathing asynchrony events can overly load the inspiratory muscles predisposing to increased work of breathing and compromised alveolar ventilation and arterial blood-gas exchange (3). We hypothesized that a computer-based breathing asynchrony algorithm may be useful for providing continuous and automatic surveillance of breathing during NIV for real time monitoring and to alert clinicians when breathing asynchrony occurs. To validate this algorithm we conducted a clinical study. Methods: With IRB approval, adults with respiratory failure and receiving NIV were studied (n= 10). Facemask pressure (Pm), flow rate, and tidal volume recordings of each patient’s breathing for approximately one minute were examined by experienced, expert critical care clinicians (n = 8) for breathing asynchrony. Using their clinical judgment, clinicians evaluated a total of 88 breathing segments and scored each segment using the following rank order scale: 1 to 1.5 = No breathing asynchrony, 2 to 2.5 = Minor breathing asynchrony, 3 to 3.5 = Moderate breathing asynchrony, and 4 to 4.5 = Severe breathing asynchrony. The breathing asynchrony algorithm, employing respiratory parameters previously determined from multiple regression analysis that were found to be significant predictors of breathing asynchrony used to derive the same rank order score. The relationship of clinician and breathing asynchrony algorithm scores were compared using a Spearman rho test; alpha was set at 0.05 for statistical significance. Results: Inspiratory positive airway pressure, expiratory positive airway pressure, and FIO2 values were 13 ± 5 cm H2O, 8 ± 4 cm H2O, and 0.38 ± 0.04, respectively. The relationship of clinician and breathing asynchrony algorithm scores was: r = 0.93 (p < 0.05) and r2 = 0.86, bias and precision values were 0 and 1.8, respectively. Conclusions: The computer-based breathing asynchrony algorithm predicted or explained 86% of the variance in the clinicians’ scores for diagnosing breathing synchrony and asynchrony. It was found to significantly predict the visual assessment of breathing asynchrony of experienced critical care clinicians to provide a valid assessment of breathing asynchrony during NIV. 1. Crit Care 2013, 17: R54; 2. Resp Care 2011, 56: 153; 3. Crit Care 2013, 17:157