nyukat/GMIC
An interpretable classifier for high-resolution breast cancer screening images utilizing weakly supervised localization
Built on ResNet-18, GMIC uses a multiple instance learning approach with global awareness to generate pixel-level saliency maps for lesion localization despite training only on image-level labels. The model achieves 28.8% fewer parameters and 4.1x faster inference than ResNet-34 while maintaining higher accuracy, with PyTorch implementation supporting both GPU and CPU inference via a configurable bash pipeline that outputs predictions and visualizations for CC/MLO mammography views.
183 stars. No commits in the last 6 months.
Stars
183
Forks
52
Language
Jupyter Notebook
License
AGPL-3.0
Category
Last pushed
Jul 25, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/nyukat/GMIC"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
nyukat/breast_cancer_classifier
Deep Neural Networks Improve Radiologists' Performance in Breast Cancer Screening
lishen/end2end-all-conv
Deep Learning to Improve Breast Cancer Detection on Screening Mammography
naomifridman/BreastDCEDL
BreastDCEDL is a deep learning–ready DCE-MRI dataset of 2,070 breast cancer patients, sourced...
LidiaGarrucho/MAMA-MIA
The MAMA-MIA Dataset: A Multi-Center Breast Cancer DCE-MRI Public Dataset with Expert Segmentations
cbailes/awesome-ai-cancer
Awesome artificial intelligence in cancer diagnostics and oncology