Skip to main content

Table 3 AUC scores in validation and test sets for models fine-tuned end-to-end with varying pre-training methods at 224 and 640 input resolutions

From: BarlowTwins-CXR: enhancing chest X-ray abnormality localization in heterogeneous data with cross-domain self-supervised learning

Model

1%

10%

100%

Barlowtwin-CXR

0.6585 (0.6527, 0.6644)

0.7756 (0.7740, 0.7772)

0.8107 (0.8095, 0.8119)

Image-Net

0.6163 (0.6087, 0.6239)

0.7168 (0.7062, 0.7274)

0.7866 (0.7833, 0.7899)

  1. Scores are presented with 95% confidence intervals
Back to article page

BMC Medical Informatics and Decision Making

ISSN: 1472-6947

Contact us