Notes

A working notebook

Papers I’ve read, methods I’m learning, books that shaped my thinking, ideas that don’t have a home yet. Some of this grows into research. Most of it doesn’t, and that’s fine. Nothing here is finished.

🌱 Seedling rough, first pass, might just be a question
🌿 Sprout revisited, starting to take shape
🌳 Evergreen fairly confident

🌱

When significance is not the same as meaning

A p-value tells you a result probably exists. It does not tell you it matters. On the gap between statistical significance and clinical meaning, and why Bayesian inference handles it more honestly.

statisticsBayesianIBDresearch methods

🌿

What Colonic Motility Actually Means (And Why That Depends on Who You Ask)

The word motility means different things depending on who you ask. What I learned after two years of working on it in UC: what it is, what disturbs it, and why measuring it is harder than it sounds.

UCmotilityMRIIBDGIQuant

🌿

The English Bias in Medical AI: What Gets Lost When Models Read Radiology Reports

Reading BABEL by R.F. Kuang made me think about how LLMs handle meaning across languages, and what that costs us in clinical settings.

AINLPradiologyclinical-AIreading-notes

🌱

Archived Notes: Anthropic’s Building Effective Agents

Notes on the spectrum between rigid clinical workflows and autonomous agents.

researchAI-agentsradiologyarchivecoding

🌱

From the Operating Room to Agentic AI: Why I Became a Clinician-Coder

Notes on going from surgery lover to agentic AI builder, and what changed in between.

personaltrajectoryreflectionradiologyAIcoding

No matching items