A Practical Gateway to Medical AI | Session 3: Spot Where Medical AI Gets It Wrong
A Practical Gateway to Medical AI | Session 3:
Spot Where Medical AI Gets It Wrong
What You’ll Get from Session 3
By the end of this 35-minute session, you can:
Explain where Medical AI may go wrong in clinical research and healthcare workflows
Recognize why LLM outputs still require clinical review, validation, and study design
Understand key concepts such as hallucination, bias and limited validation
Ask clearer questions when discussing clinical data tasks with technical collaborators
No coding background is required.
4-Session Mini-Series Roadmap
(Session 1: in person; Sessions 2–4: online; Each session is independent. Welcome to attend one, several, or all sessions.)
1: Build Your First Clinical AI Concept Map
Connect RWD, RWE, EBM, LLMs, and AI agents through one starter map.
2: Find Research Signals in Clinical Data with LLMs
Use LLMs to extract variables, cohorts, and phenotypes from clinical notes.
3: Spot Where Medical AI Gets It Wrong
See how hallucination, bias, and workflow gaps create clinical safety risks.
4: Decode AI Agents in Clinical Research Workflows in 2026
What AI agents can support, and what they cannot replace.
Who This Is For
This workshop is designed for:
Physicians · Clinicians · Trainees · Clinical Researchers · Healthcare Professionals · Students in Medicine, Public Health, Nursing, and Biomedical Fields
(No coding background required.)
Clinicians curious about AI but not looking for a heavy technical course
Clinical researchers working with patient data, EHR, notes, registries, or outcomes
Trainees, residents, fellows, and students interested in medical AI
Researchers who want to better understand AI terms used in medical papers and collaboration meetings
Anyone who wants a clearer framework for discussing clinical AI with technical teams
Format
A four‑session workshop series: Each session is independent, so you are welcome to attend one, several, or all sessions. No coding background is required.
35 minutes per session
Session 1: in person
Sessions 2–4: online
Free to attend
For Session 3
Free online workshop
35-minute concept-map talk
Post-event Q&A
No coding required
Time and Location
Date: June 18 2026
Time: 10:00 AM-10:40 AM ET
Talk: 10:00 AM-10:35 AM ET
Schedule:
10:00 AM — 35-minute talk
10:35 AM — Q&A
Participant Bonus:
As part of registration, you may submit one clinical question, data task, or research idea for AI-supported guidance. Selected submissions may receive a brief follow-up guide after the workshop with suggested AI tools, workflow ideas, and practical setup directions. Please do not include any patient-identifiable information or sensitive patient data.
Community Support :
The Harvard Medical School Chinese Scholars and Scientists Association (HMSCSSA)
Notice:
By attending this event, you consent to being photographed or recorded for documentation and promotional purposes. Informational only. Speakers’ views are their own and do not represent the views of Harvard Medical School.
Public event information may be shared with attribution. Workshop slides, handouts, concept maps, recordings, and teaching materials are created by Litong Jiang for this clinical AI workshop series. They may not be reproduced, adapted, uploaded, or reused for another course, workshop, or commercial purpose without prior written permission.