The Physician AI Handbook

A Practical Guide for Clinicians Across All Specialties

Evidence-based guidance for physicians on evaluating, implementing, and optimizing AI tools across all medical specialties. 30 comprehensive chapters with 82+ peer-reviewed citations from JAMA, NEJM, Lancet.
Author

Bryan Tegomoh, MD, MPH

Published

November 2025

Welcome to The Physician AI Handbook

As a physician who transitioned into medical informatics and epidemiology, I kept encountering a frustrating gap: research papers tout impressive AI metrics, vendors promise revolutionary improvements, yet practicing physicians face critical unanswered questions. Which AI tools actually work in clinical practice? How do I evaluate vendor claims against real evidence? What are the medico-legal implications when AI is wrong?

After years of reviewing medical literature, testing clinical AI tools, and synthesizing research from JAMA, NEJM, The Lancet, and Nature Medicine, I realized: if I was struggling to piece this together with formal training in both medicine and informatics, other physicians probably were too.

This handbook is the result of that synthesis work. It’s not vendor marketing or AI hype. It’s curated, evidence-based guidance designed to save you the searching I did—think of it as the clinical field guide I wished existed when I started evaluating AI for patient care.

Every chapter represents my attempt to distill hours of literature review, tool testing, and practical experimentation into what might actually be useful for clinical practice. I’m still learning, and this handbook will evolve as the field does.

Quick Start: Choose Your Path

Select the pathway that matches your specialty and immediate needs:

Primary Care & Family Medicine

“I need practical AI tools for my daily practice”

Start here (20 min):

Diagnostic Specialties

“Radiology, Pathology, Dermatology, Ophthalmology”

Start here (30 min):

Surgical Specialties

“General Surgery, Orthopedics, Neurosurgery, OBGYN”

Start here (25 min):

Medical Specialties

“Internal Medicine, Cardiology, Oncology, Neurology”

Start here (30 min):

Emergency & Critical Care

“I work in fast-paced, high-stakes clinical environments”

Start here (25 min):

Pediatrics & Neonatology

“I care for pediatric and newborn patients”

Start here (25 min):

New to AI entirely? → Begin with the Preface then Chapter 1: AI in Medicine

What is this handbook?

The Physician AI Handbook is an open-source, evidence-based practical guide for understanding and applying artificial intelligence in clinical medicine—written by a physician (MD, MPH, UC Berkeley) for physicians.

30 comprehensive chapters • 82+ peer-reviewed citations • Free and open-source

This is NOT another hype-filled “AI will revolutionize everything” book.

This is a clinical field guide for:

  • Practicing physicians across all specialties who need to understand AI tools and their limitations
  • Residents and fellows preparing for AI-augmented medical practice
  • Medical students entering a healthcare landscape transformed by AI
  • Hospital administrators making informed decisions about AI adoption
  • Clinical researchers exploring AI applications in their fields

What makes this different?

What you’ll get:

  • Evidence-based guidance with citations from JAMA, NEJM, Lancet, Nature Medicine, BMJ
  • Real clinical case studies (successes and failures)
  • Specialty-specific applications across 12+ medical disciplines
  • Honest assessments of what AI can and cannot do
  • Practical implementation guidance for your clinical workflow
  • Medical-legal considerations and liability frameworks
  • Open access forever

What you won’t get:

  • Generic AI hype without clinical evidence
  • Oversimplified “AI will replace doctors” narratives
  • Ignoring the complexity of real patients
  • Vendor marketing disguised as education
  • Theoretical concepts without practical application
  • Paywalled content or hidden fees

Book Structure: Your Roadmap

Part I: Foundations

AI history in medicine, fundamentals, clinical data challenges

Chapters 1-3 | 2-3 hours | Start here if new to AI

Key topics: Medical AI history (MYCIN to modern deep learning), AI fundamentals for clinicians, EHR data quality, clinical datasets

Part II: Clinical Specialties

AI applications across 12+ medical specialties

Chapters 4-15 | 8-10 hours | Jump to your specialty

Key topics: Radiology, Internal Medicine, Surgery, Pediatrics, OBGYN, Emergency/Critical Care, Oncology, Cardiology, Neurology, Primary Care, Pathology, Dermatology/Ophthalmology

Part III: Implementation & Evaluation

Clinical deployment, ethics, privacy, safety, liability

Chapters 16-21 | 4-5 hours | Critical for implementation

Key topics: Evaluating AI tools, medical ethics & equity, HIPAA compliance, clinical AI safety, workflow integration, medical liability & malpractice

Part IV: Practical Tools

Hands-on guidance for AI in daily practice

Chapters 22-25 | 3-4 hours | Immediately applicable

Includes: AI toolkit for physicians, LLMs in clinical practice (ChatGPT, Claude, Copilot), AI-assisted documentation (ambient scribes), clinical research with AI

Part V: The Future

Emerging technologies, policy, global health, future perspectives

Chapters 26-30 | 4-5 hours | Forward-looking insights

Topics: Emerging AI technologies, global health equity, healthcare policy & governance, medical misinformation, the physician-AI partnership

How to use this handbook

Choose Your Path

  • For specialists → Jump to your specialty chapter
  • For generalists → Start with Primary Care & Practical Tools
  • For residents/students → Read sequentially Part I → V
  • For administrators → Focus on Implementation & Ethics

About the Author

Bryan Tegomoh, MD, MPH is a physician and epidemiologist with experience spanning clinical medicine, health informatics, and disease surveillance. He earned his medical degree and practiced clinical medicine before completing his MPH at the University of California, Berkeley School of Public Health, where he focused on epidemiology and health data science.

Recognizing the transformative potential of AI in medicine—and the critical need for evidence-based physician education—Bryan invested years reviewing medical literature, testing clinical AI tools, and synthesizing research from leading journals including JAMA, NEJM, The Lancet, Nature Medicine, and specialty-specific publications.

This handbook emerged from that synthesis work: translating technical AI research into practical clinical guidance, evaluating vendor claims against peer-reviewed evidence, and organizing scattered information into a comprehensive resource specifically designed for practicing physicians who need to understand AI capabilities, limitations, and real-world applications without becoming machine learning engineers.

Acknowledgements & Inspiration

This handbook draws inspiration from excellent clinical and technical resources including:

  • The Epidemiologist R Handbook by Applied Epi
  • Stanford’s AI in Healthcare research and education programs
  • Research published in JAMA, NEJM, The Lancet, Nature Medicine, BMJ, and specialty journals
  • The open-source medical informatics and clinical AI research communities
  • AI Global Health Blog for practical AI perspectives

Nearly everything valuable here builds on published research, clinical implementations, and the work of countless physicians, researchers, and informaticists advancing medical AI. My contribution is synthesis and translation—gathering evidence, testing tools, and organizing knowledge specifically for clinical audiences. Credit for insights belongs to those whose work I learned from; responsibility for errors is mine.

Contributing

This is a living handbook. Your contributions make it better:

Share Case Studies Email your examples

Support the Project Share with colleagues

Terms of Use

License

This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0).

You are free to: Share and use this material for educational and non-commercial purposes with attribution.

Commercial use, derivatives, and adaptations require written permission. Contact for permissions

Full license details | CC BY-NC-ND 4.0 Legal Code

Citation

Tegomoh, Bryan. The Physician AI Handbook: A Practical Guide for Clinicians Across All Specialties. 2025. https://physicianaihandbook.com. Accessed [Date].

Academic & Clinical Use

Medical schools, residency programs, CME courses, and hospital training programs are welcome to use this material for non-commercial teaching purposes with proper attribution. Please let us know how you’re using it!

Your Feedback Makes This Better

I wrote this from my perspective and experience, which means it has blind spots. Contributions are welcome:

  • Have a better approach or implemented something different? Email me
  • Working in a specialty not well-covered? Help improve those chapters
  • Have clinical experience with AI tools? Share your real-world insights
  • Find this helpful? Share with colleagues

Follow updates on Twitter/X

TipReady to start?

New to AI? → Begin with the Preface then Chapter 1: AI in Medicine

Want specialty-specific guidance? → Jump to Part II: Clinical Specialties

Need practical tools now? → Check Chapter 22: AI Tools Every Physician Should Know

Evaluating a vendor’s AI product? → Read Chapter 16: Evaluating AI Clinical Decision Support