Seed Grant Profile
Evidence-Based Cost-Effective Diagnosis for Pediatric Cardiac Disease: A Machine Learning and Data Mining ApproachThe objective of this project is to explore machine learning and data mining methodologies to develop best practices for select pediatric cardiology diagnoses. The emphasis is on systematic exploration of historical clinical and claims data in order to develop cost-effective approaches for high quality clinical decision-making, and adaptive cost-effective methods for patient attribute acquisition. The goal is the most cost effective, highest quality care delivery and continued further evolution of best practices through ongoing analysis of care patterns and clinical outcomes. Our specific aim will be the outpatient evaluation of chest pain; a common complaint in outpatient pediatric cardiology clinics, though one with an infrequent association with true cardiac pathology.
Investigators: Hongyuan Zha (Georgia Institute of Technology), Patricio Frias (Sibley Heart Center at Children's Healthcare of Atlanta), Mark Braunstein (Georgia Institute of Technology), Haesun Park (Georgia Institute of Technology) and Alex Gray (Georgia Institute of Technology)