<p>Classification comparison of models trained on real versus synthetic data. Bold letters indicate improved performance. C, M, CX14, and S represent CheXpert, MIMIC-CXR, ChestX-ray14, and Synthetic datasets, simultaneously. ResNet50 and Swin-T are the CLIP-CXR encoders used for each model.</p>
收藏NIAID Data Ecosystem2026-05-10 收录
数据链接：
官方服务：
资源简介：
Classification comparison of models trained on real versus synthetic data. Bold letters indicate improved performance. C, M, CX14, and S represent CheXpert, MIMIC-CXR, ChestX-ray14, and Synthetic datasets, simultaneously. ResNet50 and Swin-T are the CLIP-CXR encoders used for each model.
创建时间：
2026-04-03
