lishen/end2end-all-conv

Deep Learning to Improve Breast Cancer Detection on Screening Mammography

51
/ 100
Established

Implements end-to-end whole-image classification using an all-convolutional architecture that converts patch-level classifiers (ResNet50, VGG16) into full mammogram predictors by appending convolutional and heatmap layers. Achieves 0.88-0.96 AUC across DDSM and INbreast datasets through two-stage training with patch sampling, patch classification, and whole-image fine-tuning. Provides pre-trained models and transfer learning utilities enabling rapid adaptation to custom mammography datasets via feature-wise centering and data augmentation.

389 stars. No commits in the last 6 months.

Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 10 / 25
Maturity 16 / 25
Community 25 / 25

How are scores calculated?

Stars

389

Forks

137

Language

Jupyter Notebook

License

Last pushed

Feb 24, 2022

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/lishen/end2end-all-conv"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.