Radiotherapy · Benchmarking platform
AccurayLab
Built and operated for a real treatment-planning competition: seven final plans were evaluated across 36 structures each and ranked on an anonymized 150-point leaderboard.
MEDICAL PHYSICS · CLINICAL SOFTWARE · 2023–2026
I am a medical physicist who builds and operates the software. This page is the short version for hiring teams and collaborators: the radiotherapy systems that reached real use, and the research behind them. Everything else I make lives on the blog.
Read the story →
Lunit Medical Foundation Model HackathonBenchmark Award · Team AIM · 1st place
Result, method & limits →
Clinical systems
One platform ran a real planning competition. One serves patients’ DICOM-RT records. One is an open planning stack. Each page states what the system does not prove.
Radiotherapy · Benchmarking platform
Built and operated for a real treatment-planning competition: seven final plans were evaluated across 36 structures each and ranked on an anonymized 150-point leaderboard.
Medical AI · Reproducible challenges
I kept wondering why a medical society should need an equipment vendor—or its own software team—to run a credible challenge. Orcaton is a working experiment in another model.
Radiotherapy · Patient tools
A stateless web service that turns a patient's DICOM-RT records into CT dose overlays, structures, DVHs, treatment summaries, and plain-language explanations.
Treatment planning · Monte Carlo
An open, reproducible radiotherapy planning stack spanning DICOM import, fluence optimization, MCO, MLC sequencing, virtual delivery, and independent photon, proton, and carbon Monte Carlo experiments.
Claude · Clinician-in-the-loop
A clinician-supervised planning workspace where an agent proposes, optimizes, compares, and revises research plans while every consequential step remains subject to human approval.
DICOM-RT · Privacy check
Inventories RT objects and flags obvious privacy or reference problems entirely in the browser, without uploading patient files anywhere.
Medical AI research
Public research notes on medical-image restoration, segmentation, radiotherapy, and DICOM.
Contact
Medical physics and medical AI by training; clinical software by practice. Seoul, Korea. For roles, research collaboration, technical consulting, or prototype review, reach me through LinkedIn or GitHub. Field notes on how these systems were built are on the Field Notes page.