Status: Funded - Open
Predictive analytics to prevent adverse events in interstage single ventricle heart disease
Animesh (Aashoo) Tandon, MD, MS
BACKGROUND: Infants with single ventricle heart disease undergo a series of surgeries in order to survive, and have high mortality between the first two stages of palliation. These patients are amongst the most fragile pediatric populations. We aim to prevent adverse events at home in this population by using predictive analytics to identify patterns in vital signs that can be monitored in real-time.
GAP: The causes of interstage mortality are not well understood, and monitoring these patients while they are at home is a complex process. We aim to use a novel wireless pulse oximeter to help understand these patients better and potentially decrease morbidity and mortality.
HYPOTHESIS: We hypothesize that the pulse oximeter device will be well-tolerated by infants; the device will not cause undue stress to caregivers; and that the device will obtain data that will be useable in future studies. We further hypothesize that predictive analytics of real-time vital sign patterns can be used to design an interstage adverse event risk score.
METHODS: This will be a prospective, longitudinal cohort study that will obtain continuous pulse oximeter and heart rate data from interstage patients, and use these data to devise an adverse event risk score using machine learning algorithms.
RESULTS: The results of this study are pending.
IMPACT: The results of this study could help us better predict when patients with interstage heart disease will have adverse events, which might allow us to intervene earlier. The results of this study may also allow