Simulation-First Interventional Pain Training
Trainees should meet the needle, the fluoroscope and the ultrasound probe on a spine they can study before they meet a patient.

Clinic Chief, Stanford Pain Management CenterClinical Associate Professor, Stanford University
Pain medicine, measured.
Teaching and choosing pain interventions on evidence, and building the statistics and AI that make the evidence trustworthy.
Background: fluoroscopy and ultrasound simulated in the Stanford Interventional Pain Atlas from TotalSegmentator CT data (CC BY 4.0).
Topics for keynotes, grand rounds, workshops and panels. Each can be tailored to the audience.
Trainees should meet the needle, the fluoroscope and the ultrasound probe on a spine they can study before they meet a patient.
PRP, bone marrow concentrate, stem cells and prolotherapy are marketed faster than they are tested. Here is what the trials show.
Patients deserve explanations of their procedure options that are accurate, sourced and in their own language.
Sham controls, responder analyses and small trials can make a procedure look better or worse than it is. A biostatistician who performs procedures explains how to read them.
Pain care improves only as fast as we can measure it. Here is how a registry of tens of thousands of patients turns questionnaires into evidence.
An AI that quotes a cutoff from the wrong population gives a confident, wrong answer. Grounding, attribution and refusal matter more than fluency.
Two strands: tools that teach pain procedures and weigh which ones deserve to be done, and the statistics and AI that make clinical measurement trustworthy.
For pain medicine trainees, faculty and clinicians
An education platform for pain medicine, led by the Interventional Pain Atlas, which offers fluoroscopy and ultrasound procedure simulation on a spine trainees can study. It also covers neuroanatomy, mechanisms, board-exam questions and a pain-science knowledge graph.
For clinicians, educators and the patients they counsel
Structured literature reviews of emerging pain interventions, turned into illustrated, bilingual (English and Traditional Chinese) guides in which every claim traces to a source. Topics include regenerative treatments, minimally invasive spine procedures, Tx360 for migraine, cryoneurolysis, pulsed radiofrequency and noninvasive neuromodulation.
For pain clinics, health systems and outcomes researchers
The Collaborative Health Outcomes Information Registry, Stanford's open-source learning health system for pain. It collects NIH PROMIS patient-reported outcomes through computerized adaptive testing at the point of care, and feeds clinical care, research and quality improvement from the same data.
For clinicians, researchers and developers of clinical AI
A psychometric interpretation service for AI systems, served over the Model Context Protocol with a web app on top. Given a measure and a raw score, it returns a severity band, and a T-score and percentile only where a defensible reference population exists, saying plainly when one does not.
For AI evaluation researchers and clinical AI developers
A benchmark of grounded psychometric reasoning in language models. It tests whether a model can retrieve a measurement fact, attribute it to the study that established it, compute with it, recognize when the fact does not exist or the question is malformed, and judge whether it applies to the patient at hand.
Ming-Chih Kao, PhD, MD, is Clinic Chief of the Stanford Pain Management Center and a Clinical Associate Professor of Anesthesiology, Perioperative and Pain Medicine at Stanford University. He works at the meeting point of two fields. As an interventional pain physician, board certified in Physical Medicine and Rehabilitation and in Pain Medicine and holding the World Institute of Pain's FIPP and CIPS credentials, he leads the Stanford Pain Atlas, which teaches procedures through fluoroscopy and ultrasound simulation, and an evidence library that weighs emerging interventions. As a biostatistician, trained at Harvard, he helped build Stanford's CHOIR learning health system and now builds MeasureGPT, which grounds AI systems in verified measurement facts, and ESTIMAND, which tests whether language models reason correctly about clinical measures. He speaks on interventional training, the evidence behind new procedures, measurement and trustworthy AI in medicine.
From more than 30 journal articles.
For speaking, include the event, date, location, audience and format. More on what to include. Bios, a headshot and key facts are in the press kit.