Research map / Detection and segmentation / CNN features
URICA: A Uniformity Region Affine Identifier Capture Algorithm for Arbitrary Region Retrieval in Pathology Images
Ri Su, Zhao Chen, Caleb Chen Cao, Lei Chen
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 |
|---|---|---|---|
| use K-Means anchor selection instead of fixed position or spectral clustering (ablation) | URICA · TCGA inter-region retrieval | accuracy-efficiency trade-off: mTPQ 24.29 (fixed) / 43.15 (spectral) -> 21.50 | Improved |
| add Bag of Shifting to spectral clustering anchor selection (ablation) | URICA with spectral clustering · TCGA inter-region retrieval | mTPQ: 43.15 -> 22.99 | Improved |
| add Bag of Shifting to K-Means anchor selection (ablation) | URICA with K-Means · TCGA inter-region retrieval | mTPQ: 21.50 -> 21.41 | No clear effect |
| add Bag of Shifting to fixed position anchors (ablation) | URICA with fixed position · TCGA inter-region retrieval | mTPQ: 24.29 -> 29.60 | Hurt |
| reduce anchor ratio to 0.2-0.4 (ablation) | URICA · TCGA region retrieval | retrieval accuracy | Hurt |
| use anchor ratio 0.6 instead of full anchors (1.0) (ablation) | URICA · TCGA region retrieval | accuracy and runtime: within 4.2% accuracy loss | Mixed |
| use tessellation step t=60 instead of smaller step t=50 (ablation) | URICA · Hematopoietic WSIs | latency and storage: 31.6% faster, 30% more storage-efficient | Improved |
| use large tessellation step t=224 (ablation) | URICA · Hematopoietic WSIs | retrieval accuracy | Hurt |
| replace slide-level and adjacent baselines with URICA tessellation plus affine identifiers | region retrieval · TCGA inter-region retrieval | mIoU@5: 0.1656 -> 0.6416 (Pul. vs Slide Method) | Improved |
| URICA versus Sample Method on semantic similarity | region retrieval · TCGA inter-region retrieval (CESC, ESCA) | mSim@5: CESC 0.8443 -> 0.6443 | Mixed |
Closest papers by meaning
- PathVQ: Reforming Computational Pathology Foundation Model for Whole Slide Image Analysis via Vector Quantization (NeurIPS 2025 · similarity 0.87)
- FOCUS: Knowledge-enhanced Adaptive Visual Compression for Few-shot Whole Slide Image Classification (CVPR 2025 · similarity 0.85)
- PathFLIP: Fine-grained Language-Image Pretraining for Versatile Computational Pathology (AAAI 2026 · similarity 0.85)
- Rethinking Transformer for Long Contextual Histopathology Whole Slide Image Analysis (NeurIPS 2024 · similarity 0.85)
- MLLM-HWSI: A Multimodal Large Language Model for Hierarchical Whole Slide Image Understanding (CVPR 2026 · similarity 0.84)
- Nonparametric Distribution Matching for Self-Supervised Whole-Slide Image Condensation (NeurIPS 2026 · similarity 0.84)
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.
