Summaries of research in medical AI and healthcare, alongside original findings.
The MIMIC-IV dataset is the quiet, unglamorous foundation behind both of my published papers. Both of my papers — the one on whether LLMs can safely explain radiology reports, and the follow-up on whether radiologists can actually catch it when they can’t — used the same underlying data source. I’ve referenced it in passing in…
Radiologist oversight of AI errors is the exact question this study set out to quantify. My last post walked through my main study on whether LLMs can safely explain radiology reports to patients — and one of the more uncomfortable findings was the calibration deficit: the models’ most dangerous errors were the ones delivered with…
This one’s personal — it’s my own peer-reviewed research, published this year in Healthcare. I’ve spent most of this blog reviewing other people’s work and other people’s technology. This time I get to walk you through mine. Here’s where it actually started. Ordinary people are already asking AI chatbots questions about their own medical information…