Patient-Specific COVID-19 Resource Utilization Prediction Using Fusion AI Model - Amara Tariq
Stanford University via YouTube
Overview
Save Big on Coursera Plus. 7,000+ courses at $160 off. Limited Time Only!
Explore a 31-minute conference talk on patient-specific COVID-19 resource utilization prediction using fusion AI models. Delve into Dr. Amara Tariq's research on efficient healthcare resource planning and allocation through machine learning techniques. Learn about the fusion modeling approach developed to predict clinical events based on patients' medical history and current indicators. Gain insights into feature importance for COVID-19 disease trajectory prediction and understand the challenges in healthcare resource management. Discover how electronic medical records (EMR) data is utilized to predict hospitalization needs. Examine the various fusion model types, including late, early, and middle fusion, and their applications in this context. Analyze the results, including calibration and race-based stratification, to better understand the model's performance and implications for future research in the field of AI-driven healthcare resource optimization.
Syllabus
Intro
Healthcare Resource Management
Problem Formulation
Cohort Selection
Data Characteristics
Challenges
Patient Information
Comorbidities
Medications
Laboratory Tests
Temporal Modeling
Fusion Models - Late Fusion
Fusion Models - Early Fusion
Fusion Models - Middle Fusion
Feature Importance
Results - Calibration
Race-based Stratification
Baseline
Taught by
Stanford MedAI