AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

RadGPT: Constructing 3D Image-Text Tumor Datasets

Pedro R.A.S. Bassi, Mehmet Can Yavuz, Ibrahim Ethem Hamamci, Sezgin Er, Xiaoxi Chen, Wenxuan Li, Bjoern Menze, Sergio Decherchi, Andrea Cavalli, Kang Wang, Yang Yang, Alan Yuille and 1 more

ICCV 2025 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 40 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
generate reports from tumor segmentation (RadGPT) instead of end-to-end VLMreport generation · UCSF external (OOD) kidney tumor >2cmsensitivity (%): 9.2 -> 97.3Improved
generate reports from tumor segmentation instead of end-to-endreport generation · UCSF external (OOD) liver tumor >2cmsensitivity (%): 19.2 -> 89.4Improved
generate reports from tumor segmentation instead of end-to-endreport generation · AbdomenAtlas 3.0 test (IID) pancreatic tumor >2cmsensitivity (%): 51.9 -> 81.5Improved
generate reports from tumor segmentation instead of end-to-endreport generation · AbdomenAtlas 3.0 test (IID) liver tumor >2cmspecificity (%): 95.9 -> 64.4Hurt
generate reports from tumor segmentation instead of end-to-endreport generation · UCSF external (OOD) pancreatic tumor <=2cmsensitivity (%): 27.5 -> 76.9Improved
adapt structured reports to hospital narrative style (RadGPT-N vs RadGPT-S) (ablation)RadGPT · UCSF external (OOD) text metricsROUGE-1: 13.56 -> 22.77Improved
narrative style versus structured style (ablation)RadGPT · UCSF external (OOD) text metricsBLEU: 0.72 -> 0.74Improved
narrative style versus structured style (ablation)RadGPT · UCSF external (OOD) text metricsRadGraph-F1: 4.15 -> 2.92Hurt
train baseline on AbdomenAtlas 3.0 reports (CT2Rep) and test OODCT2Rep · UCSF external (OOD) liver tumor >2cmsensitivity (%): 49.2 -> 0.0Hurt
use RadGPT structured reports instead of human-made reportsreport content · 90 UCSF CTsvolume and HU reported: 0% -> 100%Improved
use RadGPT structured reports instead of human-made reportsreport content · 90 UCSF CTsall tumor diameters reported: 63% -> 100%Improved
LLM labeler for tumor presence (zero-shot Llama 3.1 70B)diagnostic evaluation · 447 radiologist-checked reportsaccuracy: 96%Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy