Perspectives

Insights on medical AI, radiology, and building diagnostic tools.

Medical AI Without Coding: A Symposium Talk, Written Up

A written version of the talk "Medical AI without coding", given at the KSNR 2026 Fall Symposium in October and at the Catholic University of Korea radiology department in November. It covers the shift from chatbots to agents, what this looked like in my own research, tools clinicians have built, rules for hospital use, and how to start.

7 min read

medsci-skills 6.0: A Re-audit Found Results That Were Wrong

5 min read

Version 6.0 added almost no features. It cut the skills down and re-audited the whole toolkit, which found wrong results in sample size, de-identification, reference verification and meta-analysis. Here is what was fixed and which earlier results to re-check.

Three Papers Cite It. Nobody Has Written About Using It.

9 min read

Five months of medsci-skills in numbers: 302 stars, 73 forks, three academic citations, and zero third-party usage reports. What that gap means, why I am not publishing download counts, and why I expect the project's growth to stop.

Three Demos, Three Study Types: MedSci Skills End-to-End Pipeline

6 min read

We ran three public datasets — Wisconsin Breast Cancer, BCG vaccine meta-analysis, and NHANES obesity — through MedSci Skills. Each produced a complete manuscript, publication-ready figures, reporting compliance audit, and presentation slides.

Hello, Aperivue

1 min read

Introducing Aperivue — building smarter tools for medical imaging and radiology AI.