Research map / Detection and segmentation / Open-set and few-shot detection
Unsupervised Microvascular Image Segmentation Using an Active Contours Mimicking Neural Network
Shir Gur, Lior Wolf, Lior Golgher, Pablo Blinder
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 rank loss L_rank (ablation) | Ours · DeepVess dataset | AP: 0.100 -> 0.909 | Improved |
| add rank loss L_rank (ablation) | Ours · DeepVess dataset | DICE: 0.159 -> 0.829 | Improved |
| add active contour loss L_AC (ablation) | Ours · DeepVess dataset | AP: 0.617 -> 0.909 | Improved |
| add active contour loss L_AC (ablation) | Ours · DeepVess dataset | DICE: 0.437 -> 0.829 | Improved |
| add morphological pooling layers (S instead of S-bar) (ablation) | Ours · DeepVess dataset | DICE: 0.707 -> 0.829 | Improved |
| add morphological pooling layers (S instead of S-bar) (ablation) | Ours · DeepVess dataset | AP: 0.831 -> 0.909 | Improved |
| add remaining losses (tight, rec, MV, ME) to L_AC + L_rank (ablation) | Ours · DeepVess dataset | DICE: 0.817 -> 0.829 | Improved |
| add L_rank to L_AC alone (ablation) | Ours · DeepVess dataset | DICE: 0.798 -> 0.817 | Improved |
| unsupervised fine-tuning on test-domain data (FT) | Ours · DeepVess dataset -> VesselNN dataset | DICE: 0.662 -> 0.715 | Improved |
| unsupervised fine-tuning on test-domain data (FT) | Ours · VesselNN dataset -> DeepVess dataset | DICE: 0.803 -> 0.808 | No clear effect |
| unsupervised method instead of supervised DeepVess in cross-dataset transfer | segmentation · VesselNN dataset -> DeepVess dataset | DICE: 0.506 -> 0.803 | Improved |
| unsupervised method instead of supervised DeepVess in within-dataset training | segmentation · VesselNN dataset | DICE: 0.757 -> 0.776 | Improved |
Closest papers by meaning
- Diffusion Adversarial Representation Learning for Self-supervised Vessel Segmentation (ICLR 2023 · similarity 0.89)
- Self-Supervised Vessel Segmentation via Adversarial Learning (ICCV 2021 · similarity 0.88)
- vesselFM: A Foundation Model for Universal 3D Blood Vessel Segmentation (CVPR 2025 · similarity 0.88)
- DeNVeR: Deformable Neural Vessel Representations for Unsupervised Video Vessel Segmentation (CVPR 2025 · similarity 0.86)
- Upping the Game: How 2D U-Net Skip Connections Flip 3D Segmentation (NeurIPS 2024 · similarity 0.83)
- Rethinking U-Net: Task-Adaptive Mixture of Skip Connections for Enhanced Medical Image Segmentation (AAAI 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.
