Medical Image Retrieval on BreakHis average across all magnifications
96.7mAP@5A-ARVGAE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| A-ARVGAEFeature Extraction=UNI, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 96.7 | 91.5 | |
| A-ARVGAEFeature Extraction=BioMedCLIP, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 96.6 | 91.7 | |
| A-ARVGAEFeature Extraction=CCL, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 96.5 | 91.9 | |
| A-ARVGAEFeature Extraction=MobileNetV2, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 95 | 87.4 | |
| A-ARVGAEFeature Extraction=DenseNet121, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 94.8 | 87.4 | |
| A-ARVGAEFeature Extraction=VGG19, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 94.4 | 85.2 | |
| A-ARVGAEFeature Extraction=NASNetLarge, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 91.7 | 77.8 | |
| A-ARVGAEFeature Extraction=EfficientNetV2M, Magnification=BreakHis Avg., Retrieval Model=attention-based adversarially regularized variational graph autoencoder2024.05 | 91.1 | 79.1 |