Application of Predictive Analytics Algorithms to Reduce Mechanical Ventilation time after Cardiac Surgery

BOSTON CHILDREN’S HOSPITAL

Principal Investigator:  Daniel Lee Hames

Abstract: Patients with congenital heart disease (CHD) frequently require mechanical ventilation (MV) to facilitate recovery following surgery. Available MV weaning protocols do not consider the unique physiology of CHD patients. Any MV weaning protocol involving CHD must consider both the respiratory and non-respiratory support that these unique patients receive from MV. Our primary aim is to utilize three artificial intelligence (AI) based predictive analytics algorithms— each informed by continuous, high-fidelity hemodynamic and respiratory data—to develop and validate a clinical decision support system (CDSS) to guide MV weaning in CHD patients. We hypothesize that a CDSS powered by these novel risk analytic algorithms will reduce MV duration and associated morbidity in patients following surgery for CHD.

Our overall objective is to evaluate the impact of an AI informed CDSS in facilitating safe and timely discontinuation of MV in patients following surgery for CHD. Our central hypothesis is that such a CDSS will shorten the duration of MV in post-operative CHD children. the de- escalation of ICU therapies in post-operative CHD patients.