Research map / Detection and segmentation / Weakly supervised detection and segmentation
"Delving into Universal Lesion Segmentation: Method, Dataset, and Benchmark"
Yu Qiu, Jing Xu
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 X(6) extra layers to FCN (ablation) | FCN baseline · SegLesion val (VGG16) | DSC: 27.32 -> 28.61 | Improved |
| add KEM (ablation) | FCN + X(6) · SegLesion val (VGG16) | DSC: 28.61 -> 35.68 | Improved |
| add simple decoder fusing 1/4 and 1/8 scales (ablation) | FCN + X(6) + KEM · SegLesion val (VGG16) | DSC: 35.68 -> 39.19 | Improved |
| add deep supervision (ablation) | KEN · SegLesion val (VGG16) | DSC: 39.19 -> 41.01 | Improved |
| add ER to KEM (ablation) | KEN · SegLesion val (VGG16) | DSC: 34.73 -> 41.01 | Improved |
| use D'' knowledge embedding in addition to ER (ablation) | KEN · SegLesion val (VGG16) | DSC: 32.91 -> 41.01 | Improved |
| fuse X(3) rather than X(2) in decoder (ablation) | KEN · SegLesion val (VGG16) | DSC: 41.01 -> 39.45 | Hurt |
| fuse X(3) rather than X(4) in decoder (ablation) | KEN · SegLesion val (VGG16) | DSC: 37.79 -> 41.01 | Improved |
| increase KEM channels C from 512 to 1024 (ablation) | KEN · SegLesion val (VGG16) | DSC: 41.01 -> 40.68 | No clear effect |
| change K of KEM from 24 to 48 (ablation) | KEN · SegLesion val (VGG16) | DSC: 41.01 -> 38.40 | Hurt |
| use ResNet50 backbone instead of VGG16 | KEN · SegLesion test set | DSC: 41.01 -> 42.06 | Improved |
Closest papers by meaning
- Unleashing the Potential of Vision-Language Pre-Training for 3D Zero-Shot Lesion Segmentation via Mask-Attribute Alignment (ICLR 2025 · similarity 0.86)
- Unified Medical Lesion Segmentation via Self-referring Indicator (CVPR 2025 · similarity 0.86)
- Minding Fuzzy Regions: A Data-driven Alternating Learning Paradigm for Stable Lesion Segmentation (CVPR 2025 · similarity 0.85)
- Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.84)
- Semantic-guided Masked Mutual Learning for Multi-modal Brain Tumor Segmentation with Arbitrary Missing Modalities (AAAI 2025 · similarity 0.84)
- HACDR-Net: Heterogeneous-Aware Convolutional Network for Diabetic Retinopathy Multi-Lesion Segmentation (AAAI 2024 · 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.
