Research map / Label-efficient and robust learning / Recognition and long-tail
Structure-Leveraged Methods in Breast Cancer Risk Prediction
Jun Fan, Yirong Wu, Ming Yuan, David Page, Jie Liu, Irene M. Ong, Peggy Peissig, Elizabeth Burnside
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add group and l_p fusion structure penalties to logistic regression | logistic regression · Marshfield breast cancer, Mammo model | AUC: 0.700 -> 0.723 | Improved |
| add group and l_p fusion structure penalties to logistic regression | logistic regression · Marshfield breast cancer, SNP77 model | AUC: 0.590 -> 0.684 | Improved |
| add group and l_p fusion structure penalties to logistic regression | logistic regression · Marshfield breast cancer, Combined model | AUC: 0.697 -> 0.766 | Improved |
| add structure penalties on top of lasso | LR+Lasso · Marshfield breast cancer, Combined model | AUC: 0.721 -> 0.766 | Improved |
| add SNPs to mammography features without regularization | logistic regression · Marshfield breast cancer | AUC: 0.700 -> 0.697 | No clear effect |
| add SNPs to mammography features with structure penalties | LR+Structure · Marshfield breast cancer | AUC: 0.723 -> 0.766 | Improved |
| use fusion penalty instead of group penalty | LR on SNP77 · Marshfield breast cancer, SNP77 model | AUC: 0.614 -> 0.676 | Improved |
| use group penalty instead of fusion penalty | LR on mammography · Marshfield breast cancer, Mammo model | AUC: 0.710 -> 0.716 | Improved |
| use p=1 rather than p=2 fusion shrinkage (ablation) | fused group lasso · Marshfield breast cancer, Combined model | AUC: 0.745 -> 0.757 | Improved |
| use p1=2 for mammography and p2=1 for SNPs rather than p1=p2=1 (ablation) | fused group lasso (method 3) · Marshfield breast cancer, Combined model | AUC: 0.757 -> 0.766 | Improved |
| use separate p (p1=2, p2=1) rather than p1=p2=2 (ablation) | fused group lasso (method 3) · Marshfield breast cancer, Combined model | AUC: 0.745 -> 0.766 | Improved |
| increase p2 for SNP group fusion (ablation) | fused group lasso (method 3) · Marshfield breast cancer, Combined model | AUC: decreases as p2 increases | Hurt |
Closest papers by meaning
- Factorized Context Aggregation for Robust Cancer Risk Estimation via Soft Re-Ranked Retrieval and Hierarchical Anchors (CVPR 2026 · similarity 0.85)
- Learning the Finer Things: Bayesian Structure Learning at the Instantiation Level (AAAI 2023 · similarity 0.85)
- Incorporating Grouping Information into Bayesian Decision Tree Ensembles (ICML 2019 · similarity 0.84)
- Bayesian group factor analysis with structured sparsity (JMLR 2016 · similarity 0.84)
- PersonalizedUS: Interpretable Breast Cancer Risk Assessment with Local Coverage Uncertainty Quantification (AAAI 2025 · similarity 0.82)
- Dual Feature Reduction for the Sparse-group Lasso and its Adaptive Variant (ICML 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
