Meet the speakers, judges and advisors at the Health AI Symposium >
| PI/Lead Innovator | Title | Hospital Affiliation | Innovation name | Description | |
|---|---|---|---|---|---|
 |
Associate Professor of Surgery (Transplant), Yale School of Medicine | Yale New Haven Health | Deep Machine Learning Model for Prediction of Death in Organ Donation after Circulatory Death (DCD) | AI-enabled deep machine learning model that predicts time-to-death following terminal extubation in DCD donors leveraging dynamic longitudinal clinical data using an ODE-RNN framework to optimize organ procurement processes, reduce dry runs, and enhance transplant outcomes. | |
 |
David Chen, MD, FAAPMR, MMSc | Assistant Professor of Medicine (Cardiovascular Medicine) and of Biostatistics (Health Informatics) | Gaylord Specialty Care | GaylordAuth AI, an AI-Driven Pre-Authorization Optimization Platform | AI-driven pre-authorization optimization platform for LTACHs that leverages advanced NLP and machine learning to predict authorization needs, automate form completion, provide real-time denial risk scoring, dynamically map payer policies, and generate intelligent appeal drafts. It streamlines the prior authorization process to reduce denials and administrative workload while enhancing revenue cycle efficiency. |
 |
Tze Chiam, PhD | Senior Director, Health Systems Analytics and Modeling | Connecticut Children’s Assistant Professor, Department of Pediatrics, UCONN School of Medicine |
Connecticut Children’s | The Use of AI, Computer Simulation and Mathematical Optimization for Optimizing Emergency Department Resources Allocation | Cloud-based ED operations management platform that integrates AI/ML forecasting, optimization modeling, discrete-event simulation, and digital twin technology. It dynamically predicts patient volumes, optimizes staffing, and enables real-time scenario analysis to proactively and agilely respond to overcrowding without disrupting emergency care. |
 |
Alex Hogan, MD, MS | Professor, Pediatrics | Connecticut Children’s | Automation of Low-Value Care Metrics Using Natural Language Processing | NLP-powered automation platform for real-time measurement of low-value care to extract key clinical features from unstructured EHR notes, leveraging advanced language models to eliminate manual chart review and drive cost-effective quality improvement in pediatric asthma care. |
 |
Rohan Khera, MD, MS | Assistant Professor of Medicine (Cardiovascular Medicine) and of Biostatistics (Health Informatics), Yale School of Medicine | Yale New Haven Health | Ensight-AI | AI-enabled ECG analysis tool that uses deep learning on standard ECG images to detect subclinical myocardial diseases—such as left ventricular systolic dysfunction and amyloid cardiomyopathy—in real time, facilitating early diagnosis and timely intervention. |
 |
Po-Jen Lin, MD, MPH | Resident | Nuvance Health | Re-Agent | AI-driven research collaboration platform that employs agentic AI to automate complex administrative workflows—from IRB applications to funding proposals. It leverages blockchain technology for secure, transparent data sharing and contribution tracking. This innovative solution streamlines cross-institutional population health research, reducing bureaucratic overhead and fostering equitable, efficient collaboration on a global scale. |
 |
Changchun Liu, PhD | Professor, Biomedical Engineering |
UConn Health | AI-Driven Mobile Health Tool for Infectious Disease Detection at the Point of Care |
AI-powered smartphone diagnostic platform designed for rapid and accurate detection of infectious diseases at the point-of-care. Uses deep learning to enhance CRISPR-reaction efficiency and CRISPR-based fluorescence detection, enabling ultrafast and precise pathogen identification without the need for expensive optical detectors. |
 |
Tapan Mehta, MD | Director of Research and Innovation, Stroke and Cerebrovascular Diseases | Hartford HealthCare | Detection and Localization of Subdural Hematoma with Subsequent Recurrence Prediction using AI | AI-driven platform that facilitates rapid identification and precise localization of subdural hematomas. It integrates convolutional neural networks and advanced 3D segmentation models trained on brain CT scans, paired with patient-specific data including demographics and medical history. |
 |
Richa Sharma, MD, MPH | Assistant Professor, Yale School of Medicine | Yale New Haven Health | POLARIS | AI-driven co-pilot that utilizes machine learning to classify stroke etiology and extract actionable insights from unstructured EHR data, aiming to improve stroke outcomes and brain health. |
 |
David Sink, MD | Regional Clinical Director, Neonatology | Connecticut Children’s | NEC Detect: AI-Driven Early Detection Model for Severe Necrotizing Enterocolitis in Premature Infants | AI-driven NEC detection platform for preterm infants that uses advanced deep learning architectures to analyze serial abdominal radiographs and ultrasound images. The platform integrates imaging with clinical and demographic data for early, interpretable, and precise differentiation between medical and surgical NEC management. |
 |
Hua Xu, PhD | Robert T. McCluskey Professor of Biomedical Informatics and Data Science, Yale School of Medicine | Yale New Haven Health | Agentic AI for Healthcare Scheduling - A Use Case on Optimizing Anesthesiology Staff Scheduling in Surgery Rooms | AI-enabled multi-agent scheduling system that optimizes clinical resource allocation by forecasting staffing needs, generating block schedules, and dynamically adjusting assignments in real time that streamlines the entire scheduling pipeline for the anesthesiology department to enhance efficiency, reduce overtime, and improve patient care delivery. |
 |
Steven Zweibel, MD, FACC, FHRS, CCDS | Medical Director of Innovation of the Heart and Vascular Institue at Hartford Healthcare, Director of the Cardiac Device Program at Hartford Healthcare | Hartford HealthCare | AI-Enabled Multi-Class Diagnosis of Ejection Fraction from Electrocardiograms | AI-enabled multi-class diagnosis tool for ejection fraction from electrocardiograms. This tool leverages deep convolutional and recurrent neural networks in ECG tracings and patient electronic health records, integrating into the EHR at multiple sites. |
