Resources

Tools, frameworks, reading lists, and reference materials for working with AI in clinical practice.

Key Frameworks

These frameworks are referenced throughout the course and serve as practical tools for evaluating and governing clinical AI.

CLAIM Checklist
42-item checklist for evaluating AI imaging papers. Covers study design, data quality, model development, validation, and reproducibility.
Read paper (Radiology: AI, 2020) →
CONSORT-AI Extension
Reporting guidelines for clinical trial reports involving AI interventions. Essential for evaluating research quality.
View Guidelines →
DEFT-AI Supervision Framework
Diagnosis, Evidence, Feedback, Teaching—structured approach to supervising AI use by medical learners. Abdulnour, Gin, and Boscardin, NEJM 2025.
Read paper →
Model Card Template
Documentation framework for AI system transparency. Based on Mitchell et al. (2019).
Read paper →

AI Tools for Clinicians

These tools are referenced in the course and may be useful for clinical AI exploration and practice.

Tool Use Case Key Considerations
NotebookLM Document-grounded AI assistant for summarizing guidelines, papers Limited to uploaded sources; useful for controlled exploration
Claude General-purpose AI assistant with strong reasoning Check institutional policies; avoid PHI in consumer version
GPT-4 General-purpose AI with medical knowledge May hallucinate; verify all clinical information
Perplexity AI search with citations Good for research; citations help verification
Elicit Research assistant for literature review Helpful for rapid evidence synthesis
PHI Warning

Never enter protected health information (PHI) into consumer AI tools unless your institution has a BAA in place. This includes patient names, dates, medical record numbers, and any identifiable clinical details.

Professional Guidelines

AMA: Augmented Intelligence in Health Care (H-480.940)
American Medical Association's policy principles for AI in clinical practice.
View Policy →
FDA: Artificial Intelligence and Machine Learning in Software as a Medical Device
Regulatory framework for AI-based medical devices.
View Guidance →
WHO: Ethics and Governance of Artificial Intelligence for Health
World Health Organization guidance on ethical AI deployment in healthcare.
View Report →

Complete Reading List

All required and optional readings from the course, organized by topic.

Foundations

Generative AI

Governance

Ongoing Learning

Stay current with these resources:

Glossary

AI Triad

The three fundamental components of any AI system: data (what it learns from), algorithms (how it learns), and compute (the processing power that enables learning).

Automation Bias

The tendency to over-rely on automated systems and accept their outputs uncritically, even when contradicted by other information.

Centaur

A human-AI collaboration model with clear division of labor—human handles strategy and judgment, AI handles calculation and data processing.

Chain-of-Thought Prompting

A prompting technique that asks AI to show its reasoning step-by-step before providing a final answer, making its logic visible for evaluation.

Cyborg

A human-AI collaboration model with seamless integration, where the boundary between human and machine cognition is blurred.

External Validation

Testing an AI model on data from a different institution or population than where it was developed.

Few-Shot Prompting

Providing examples of desired input-output pairs before asking the AI to perform a task.

Hallucination

When an AI generates plausible but factually incorrect information.

Label Bias

Bias introduced through the outcome variable being predicted—when the "ground truth" itself encodes discrimination.

Proxy Variable

A measurable variable used as a stand-in for something difficult to measure directly—can encode hidden biases.

Shadow AI

Use of AI tools outside institutional oversight and governance, often consumer products used for clinical work.

Temporal Validation

Testing an AI model on data from a different time period than the training data.

Zero-Shot Prompting

Asking an AI to perform a task without providing examples—relying on its pre-training.