Liam Ghiam, Clinical Associate in the UCI Health, Integrative and Functional Medicine Fellow at the Susan Samueli Integrative Health Institute, shared on LinkedIn:
“This week, AI found something in a patient’s chart that I did not find myself. The patient and family did not remember either, even when I asked directly.
I was seeing a patient with metastatic disease in a neck lymph node with unknown primary site. We had done a detailed history, PE and chart review, but one important piece was still missing.
Then our AI tool, integrated into the EHR, picked up something interesting.
Buried among years of medical records was a short note from a PA documenting a prior skin cancer that had been removed.
The patient had not mentioned this history, there was no skin lesion on exam, and there was no pathology report available in the chart.
The patient was specifically and directly asked about any history of skin cancer, but neither the patient nor the family remembered this prior diagnosis.
This small piece of information could be very important. That prior skin cancer could potentially be the source of the metastatic lymph node and could change how we approach the diagnosis and treatment.
I have been a true believer in the role of AI in medicine. I know the published data are still evolving and sometimes mixed.
A recent 2026 study of an EHR-integrated AI chart review tool showed overall positive physician feedback and supported its use in clinical workflows.
Experiences like this show that AI has real value in our daily practice and can ultimately improve clinical outcomes in the real world.
In a busy clinic, it is impossible to read every note from years of a patient’s medical history. Even when we take a careful history and ask specific questions, patients and their families may not remember something that could be important.
AI can augment what we do and help us to make better and faster clinical decisions.
Sometimes the value of AI is just finding one important report that could easily get lost among thousands of notes and other data.”
Title: Evaluation of electronic health record-integrated artificial intelligence chart review
Authors: Nicolas M. Kahl, Marshall J. Frieden, Zach R. Pope, Marlene M. Millen, Vaishal M. Tolia, Theodore C. Chan, Christopher A. Longhurst, Karandeep Singh, Alan X. You
Other articles from Liam Ghiam on OncoDaily.