AI for Clinical Medicine

Intelligent Healthcare

Developing AI tools that seamlessly integrate into clinical workflows to enhance diagnosis, treatment decisions, and patient outcomes through predictive modeling and intelligent decision support.

Transforming Clinical Practice

Our clinical AI research spans the full spectrum of healthcare applications, from early prediction of clinical events to real-time decision support systems that assist clinicians at the point of care.

We focus on developing robust, interpretable AI systems that can handle the complexity and variability of real-world clinical data while maintaining the trust and confidence of healthcare providers.

1987
Temporal Reasoning for Medical Expert Systems
500+
Clinical AI publications

Key Impact Areas

  • Using Electronic Health Record data for observational trials.
  • Predictive modeling for clinical events
  • Rare Disease Management
  • AI for diagnostics and personal control

Research Focus Areas

Our multidisciplinary approach to clinical AI innovation

Predictive Modeling

Advanced machine learning models that predict clinical events, patient outcomes, and disease progression using longitudinal electronic health record data.

  • • Machine learning for temporal prediction
  • • Risk stratification and early warning systems
  • • Multimodal data fusion techniques

Clinical Decision Support

Intelligent systems that provide evidence-based recommendations to clinicians at the point of care, integrated seamlessly into existing clinical workflows.

  • • Real-time diagnostic assistance
  • • Treatment recommendation engines
  • • Clinical guideline automation

Alignment for Clinical Care

We trust our doctors to have patients' best interests foremost in guiding their decisions. How can we trust AI?                        

  • • Human values in clinical decisions.
  • • Evaluating LLMs for alignment with humans
  • • Aligning LLM's to diverse human values.   

Rare Disease Diagnostics

Helped build the national infrastructure for the Undiagnosed Disease Network with 'omic diagnoses at scale'

  • • AI in interpretation of genomes.
  • • Secure sharing across clinics.
  • • AI for rare disease diagnostics.

AI Safety & Robustness

Ensuring AI systems are robust, reliable, and safe for clinical deployment through rigorous testing and adversarial analysis.

  • • Adversarial attack detection and mitigation
  • • Model uncertainty quantification
  • • Distribution shift handling
  • Temporal Bias in prediction

Clinical Implementation

Translating research innovations into practical clinical tools that can be deployed in real healthcare settings with measurable impact on patient care.

Highlighted Publications

Influenced the implementation and design of clinical AI

Foundational Re-Framing 2009

No small change for the health information economy

How do we go from monolithic EHR applications to a flexible open market app economy. This publication led to the definition of the SMART-on-FHIR API that was adopted by most EHR vendors and Apple Health.   

EHR Architecture  Patient control 
SMART-on-FHIR API  
Impact
Recent Impact 2023

Artificial Intelligence in Medicine

Comprehensive editorial in NEJM that launched the NEJM AI journal and provided a roadmap for the future of artificial intelligence applications in clinical medicine.

Editorial Leadership NEJM AI Clinical Impact
Future Framing
Impact
Safety Research 2018

Adversarial Attacks Against Medical Deep Learning Systems

Identifying vulnerabilities in medical AI systems and establishing frameworks for ensuring robustness and safety in clinical AI deployments.

AI Safety Robustness Security
500+
Citations
Caution from Clinical Context 2009

Biases in electronic health record data due to processes within the healthcare system: retrospective observational study

The context under which data is obtained strongly influences its actionability and interpretation.   

EHR Context  Interpretation 
Clinical data and context  
Warning

Join the Future of Clinical AI

Interested in pushing the boundaries of AI in healthcare? Explore opportunities to collaborate with our team or learn more about our ongoing research.