Research map / Graph learning / Graph self-supervised learning
Deep Lesion Graphs in the Wild: Relationship Learning and Organization of Significant Radiology Image Findings in a Diverse Large-Scale Lesion Database
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 location and size cues to type-supervised triplet learning (ablation) | Lesion embeddings · DeepLesion test set | type retrieval ARE (%): 8.8±0.2 -> 8.5±0.1 | Helped |
| Add the size cue to type-and-location triplet learning (ablation) | Lesion embeddings · DeepLesion test set | size retrieval ARE (%): 6.0±0.1 -> 5.1±0.0 | Helped |
| Add fine-level multi-scale features to conv5 features (ablation) | Lesion embeddings · DeepLesion test set | type retrieval ARE (%): 11.5±0.2 -> 8.5±0.1 | Helped |
| Add fine-level multi-scale features to conv5 features (ablation) | Lesion embeddings · DeepLesion test set | location retrieval ARE (%): 7.1±0.1 -> 7.2±0.0 | No clear effect |
| Add iterative pseudo-label refinement (ablation) | Lesion embeddings · DeepLesion test set | clustering NMI (%): 69.8±2.0 -> 72.4±4.6 | Helped |
| Add iterative pseudo-label refinement (ablation) | Lesion embeddings · DeepLesion test set | type retrieval ARE (%): 8.7±0.2 -> 8.5±0.1 | No clear effect |
| Use triplet embeddings instead of multi-scale ImageNet features | Lesion retrieval · DeepLesion test set | type retrieval ARE (%): 15.2 -> 8.5±0.1 | Helped |
| Use triplet embeddings instead of location alone | Lesion retrieval · DeepLesion test set | location retrieval ARE (%): 2.5 -> 7.2±0.0 | Hurt |
| Use appearance-based triplet embeddings instead of location alone | Intra-patient lesion matching · DeepLesion | AUC | Helped |
Closest papers by meaning
- Holistic and Comprehensive Annotation of Clinically Significant Findings on Diverse CT Images: Learning From Radiology Reports and Label Ontology (CVPR 2019 · similarity 0.91)
- Large-scale and Fine-grained Vision-language Pre-training for Enhanced CT Image Understanding (ICLR 2025 · similarity 0.84)
- EasyLens: A Training-Free Plug-and-Play Subtle-Lesion Representation Amplifier for Medical Vision-Language Models (NeurIPS 2026 · similarity 0.84)
- Unified Medical Lesion Segmentation via Self-referring Indicator (CVPR 2025 · similarity 0.84)
- Minding Fuzzy Regions: A Data-driven Alternating Learning Paradigm for Stable Lesion Segmentation (CVPR 2025 · similarity 0.83)
- AHIVE: Anatomy-aware Hierarchical Vision Encoding for Interactive Radiology Report Retrieval (CVPR 2024 · similarity 0.83)
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.