Aperivue Blog
Insights on medical AI, radiology, and building diagnostic tools.
medsci-skills
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.
LI-RADS
Learn how ACR LI-RADS CT/MRI v2018 categorizes liver observations in patients at high risk for hepatocellular carcinoma. Free online calculator with structured reporting.
Lung-RADS
Learn how to apply ACR Lung-RADS v2022 to low-dose CT lung cancer screening. Free online calculator with structured reporting for solid, part-solid, and ground-glass nodules.
O-RADS
Learn how O-RADS US v2022 and O-RADS MRI 2022 score ovarian and adnexal masses. Free online calculator with lexicon-based scoring and structured reporting.
PI-RADS
Learn how PI-RADS v2.1 scores prostate multiparametric MRI, zone by zone. Free structured report generator with DWI-dominant PZ, T2W-dominant TZ, and DCE upgrade logic.
RADS
An overview of the six ACR RADS systems — TI-RADS, Lung-RADS, BI-RADS, LI-RADS, PI-RADS, and O-RADS — with a free structured report calculator for each.
medsci-skills
v2.1.0 grows to 36 skills, introduces Reference Safety Phase 1 (SSOT.yaml + verify-refs hook modes), four meta-analysis failure-mode automation scripts, and a no-terminal classroom installer bundle.
medsci-skills
MedSci Skills grows from 22 to 32 skills. Five new skills — humanize, author-strategy, peer-review, ma-scout, lit-sync — plus anti-hallucination citation verification now active across the full bundle, and a new CI validation script.
medsci-skills
MedSci Skills v2.2 introduces autonomous end-to-end manuscript generation — from raw data to submission-ready DOCX with no human gates. New Phase 7 QC chain, 33 reporting guidelines, structured JSON outputs, and post-skill validation.
BI-RADS
Learn how to use the ACR BI-RADS 5th edition lexicon for mammography, ultrasound, and MRI. Free structured report generator with PACS-ready output, kinetic curve analysis, and multi-finding support.
medsci-skills
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.
claude-code
A radiology resident with no coding background built an open-source toolkit that automates systematic reviews, manuscript writing, and statistical analysis — using nothing but Claude Code and domain expertise.
STROBE
Learn how to audit your observational study manuscript against all 22 STROBE checklist items using a free, open-source AI tool. Step-by-step guide with examples.
TI-RADS
Learn how to use ACR TI-RADS, K-TIRADS, and EU-TIRADS scoring systems for thyroid nodule assessment. Free online calculator with structured reporting.
announcement
Introducing Aperivue — building smarter tools for medical imaging and radiology AI.
medical-ai
A radiology clinic in Busan, South Korea, has built a fully AI-integrated breast imaging workflow. What does this tell us about the future of clinical AI adoption?
deep-learning
AI systems keep claiming to outperform radiologists, but what does 'better' actually mean in clinical practice? A radiologist's perspective on the nuances that benchmarks miss.
opinion
What medical training reveals about the nature of expertise in complex systems — and why formulas fail in both medicine and investing.