Deep Neural Networks With Region-Based Pooling Structures for Mammographic Image Classification

Open paper
Benchmark Model Rank Results
suspicious-birads-4-5-no-suspicious-birads-1-2-3-per-image-classification-on-inbreastAvg pooling DenseNet-169–AUC: 0.862
suspicious-birads-4-5-no-suspicious-birads-1-2-3-per-image-classification-on-inbreastGGP DenseNet-169–AUC: 0.922
suspicious-birads-4-5-no-suspicious-birads-1-2-3-per-image-classification-on-inbreastMax pooling DenseNet-169–AUC: 0.837
suspicious-birads-4-5-no-suspicious-birads-1-2-3-per-image-classification-on-inbreastRGP DenseNet-169–AUC: 0.934

Rank counts only results with a code link.