List your company
ResearchFormerly WizWand

Object Detection on RSNA

31.1mAP (%)Med-UniC
2.59.92517.3524.775May 31, 2023Nov 17, 2023May 5, 2024Oct 22, 2024Apr 10, 2025Sep 27, 2025Mar 17, 2026
Updated 7mo ago

Evaluation results

MethodLinks
Med-UniCTraining Data Fraction=100%, Disease-level annotations=false2023.0531.1
SISTAFine-tuning percentage=100%2026.0127.4
G2D (encoder)Training data fraction=100%, architecture=encoder2023.1227.2
MLIPFine-tuning percentage=100%, Auxiliary knowledge=true2026.0125.8
M-FLAGLabeled data fraction=100%2023.0725.4
M-FLAGTraining data fraction=100%2023.1225.4
M-FLAGFine-tuning percentage=100%2026.0125.4
PRIORFine-tuning percentage=100%2026.0125.2
MGCALabeled data fraction=100%2023.0724.9
MGCATraining Data Fraction=100%, Disease-level annotations=true2023.0524.9
MGCATraining data fraction=100%2023.1224.9
MGCAFine-tuning percentage=100%2026.0124.9
GLoRIALabeled data fraction=100%2023.0724.8
GLoRIA-MIMICTraining Data Fraction=100%, Disease-level annotations=false2023.0524.8
GLoRIA-MIMICTraining data fraction=100%2023.1224.8
MedKLIPTraining Data Fraction=100%, Disease-level annotations=true2023.0524.5
MedKLIP*Training data fraction=100%, annotations=disease-level2023.1224.5
MedKLIPFine-tuning percentage=100%, Instance-level labels=true2026.0124.5
SISTAFine-tuning percentage=10%2026.0124.3
GLoRIAFine-tuning percentage=100%2026.0123.1
Med-UniCTraining Data Fraction=10%, Disease-level annotations=false2023.0522.3
SISTAFine-tuning percentage=1%2026.0122
G2D (encoder)Training data fraction=10%, architecture=encoder2023.1221.7
ACE-LoRAmode=Zero-shot2026.0321.29
GLoRIAmode=Zero-shot2026.0320.42
MGCAmode=Zero-shot2026.0320.3
BiomedCLIPmode=Zero-shot2026.0320.25
BioVILmode=Zero-shot2026.0319.31
MLIPFine-tuning percentage=10%, Auxiliary knowledge=true2026.0119.1
PRIORmode=Zero-shot2026.0318.89
GLoRATraining Data Fraction=100%, Disease-level annotations=false2023.0518.8
GLoRATraining data fraction=100%2023.1218.8
PRIORFine-tuning percentage=10%2026.0118.5
KADmode=Zero-shot2026.0318.1
ConVIRTLabeled data fraction=100%2023.0717.9
ConVIRTTraining Data Fraction=100%, Disease-level annotations=false2023.0517.9
ConVIRTTraining data fraction=100%2023.1217.9
ConVIRTFine-tuning percentage=100%2026.0117.9
M-FLAGLabeled data fraction=10%2023.0717.5
M-FLAGTraining data fraction=10%2023.1217.5
M-FLAGFine-tuning percentage=10%2026.0117.5
MLIPFine-tuning percentage=1%, Auxiliary knowledge=true2026.0117.2
MGCALabeled data fraction=10%2023.0716.8
MGCATraining Data Fraction=10%, Disease-level annotations=true2023.0516.8
MGCATraining data fraction=10%2023.1216.8
MGCAFine-tuning percentage=10%2026.0116.8
Med-UniCTraining Data Fraction=1%, Disease-level annotations=false2023.0516.6
MedKLIPTraining Data Fraction=10%, Disease-level annotations=true2023.0516.3
MedKLIP*Training data fraction=10%, annotations=disease-level2023.1216.3
MedKLIPFine-tuning percentage=10%, Instance-level labels=true2026.0116.3
GLoRIALabeled data fraction=10%2023.0716.1
GLoRIA-MIMICTraining Data Fraction=10%, Disease-level annotations=false2023.0516.1
GLoRIA-MIMICTraining data fraction=10%2023.1216.1
G2D (encoder)Training data fraction=1%, architecture=encoder2023.1215.9
ImageNetLabeled data fraction=100%2023.0715.7
ImageNetTraining Data Fraction=100%, Disease-level annotations=false2023.0515.7
ImageNet InitTraining data fraction=100%2023.1215.7
ImageNetFine-tuning percentage=100%2026.0115.7
ConVIRTLabeled data fraction=10%2023.0715.6
ConVIRTTraining Data Fraction=10%, Disease-level annotations=false2023.0515.6
ConVIRTTraining data fraction=10%2023.1215.6
ConVIRTFine-tuning percentage=10%2026.0115.6
GLoRIAFine-tuning percentage=10%2026.0115.6
PRIORFine-tuning percentage=1%2026.0115.6
GLoRATraining Data Fraction=10%, Disease-level annotations=false2023.0514.8
GLoRATraining data fraction=10%2023.1214.8
M-FLAGLabeled data fraction=1%2023.0713.7
M-FLAGTraining data fraction=1%2023.1213.7
M-FLAGFine-tuning percentage=1%2026.0113.7
MGCALabeled data fraction=1%2023.0712.9
MGCATraining Data Fraction=1%, Disease-level annotations=true2023.0512.9
MGCATraining data fraction=1%2023.1212.9
MGCAFine-tuning percentage=1%2026.0112.9
GLoRIALabeled data fraction=1%2023.0711.6
GLoRIA-MIMICTraining Data Fraction=1%, Disease-level annotations=false2023.0511.6
GLoRIA-MIMICTraining data fraction=1%2023.1211.6
GLoRIAFine-tuning percentage=1%2026.0110.3
GLoRATraining Data Fraction=1%, Disease-level annotations=false2023.059.8
GLoRATraining data fraction=1%2023.129.8
RandomLabeled data fraction=100%2023.078.9
RandomTraining Data Fraction=100%, Disease-level annotations=false2023.058.9
MedKLIPTraining Data Fraction=1%, Disease-level annotations=true2023.058.9
Random InitTraining data fraction=100%2023.128.9
MedKLIP*Training data fraction=1%, annotations=disease-level2023.128.9
RandomFine-tuning percentage=100%2026.018.9
MedKLIPFine-tuning percentage=1%, Instance-level labels=true2026.018.9
ConVIRTLabeled data fraction=1%2023.078.2
ConVIRTTraining Data Fraction=1%, Disease-level annotations=false2023.058.2
ConVIRTTraining data fraction=1%2023.128.2
ConVIRTFine-tuning percentage=1%2026.018.2
ImageNetLabeled data fraction=10%2023.078
ImageNetTraining Data Fraction=10%, Disease-level annotations=false2023.058
ImageNet InitTraining data fraction=10%2023.128
ImageNetFine-tuning percentage=10%2026.018
RandomLabeled data fraction=10%2023.074
RandomTraining Data Fraction=10%, Disease-level annotations=false2023.054
Random InitTraining data fraction=10%2023.124
RandomFine-tuning percentage=10%2026.014
ImageNetLabeled data fraction=1%2023.073.6
ImageNetTraining Data Fraction=1%, Disease-level annotations=false2023.053.6