AI Paper Atlas

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

CancerUniT: Towards a Single Unified Model for Effective Detection, Segmentation, and Diagnosis of Eight Major Cancers Using a Large Collection of CT Scans

Jieneng Chen, Yingda Xia, Jiawen Yao, Ke Yan, Jianpeng Zhang, Le Lu, Fakai Wang, Bo Zhou, Mingyan Qiu, Qihang Yu, Mingze Yuan, Wei Fang and 13 more

ICCV 2023 · Detection and segmentation › Open-set and few-shot detection · approach: Interactive and referring segmentation · 26 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
add query hierarchy (detection-to-diagnosis projection) vs plain queries (ablation)CancerUniT · 8-cancer CT test setaverage detection sensitivity (%): 89.5 -> 93.3Improved
link detection and diagnosis queries hierarchically instead of parallel groups (ablation)CancerUniT · 8-cancer CT test settumor Dice: 0.608 -> 0.628Improved
link detection and diagnosis queries hierarchically instead of parallel groups (ablation)CancerUniT · 8-cancer CT test setnormal-control specificity (%): 78.9 -> 81.7Improved
train one unified model for all organs instead of 8 separate nnUNetsnnUNet · 8-cancer CT test setspecificity on normal controls (%): 59.2 -> 78.9Improved
train one unified nnUNet instead of separate per-organ modelsnnUNet · 8-cancer CT test setpatient-level diagnosis sensitivity (%): 68.1 -> 64.2Hurt
unified multi-organ training for breast tumor detectionnnUNet · 8-cancer CT test setbreast detection sensitivity (%): 96.0 -> 90.0Hurt
add Transformer query decoder with hierarchy to unified nnUNetunified nnUNet · 8-cancer CT test setdiagnosis sensitivity (%): 64.2 -> 70.9Improved
use semantic segmentation instead of object detection (LENS) for tumor detectiontumor detection · 8-cancer CT test setaverage detection sensitivity (%): 72.3 -> 88.0Improved
train on the 8-cancer dataset rather than DeepLesionlesion detection · normal-control specificityaverage specificity (%): 4.2 -> 64.8Improved
keep CancerUniT instead of TransUNet for class-agnostic lesion recallCancerUniT · 8-cancer CT test setlesion recall (%): 77.9 -> 75.0Hurt
unified CancerUniT instead of 8 single-organ models: speed and parametersCancerUniT · Inference efficiencyinference time: 187s -> 42sImproved

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