Research map / Detection and segmentation / CNN features
Shape-Aware Organ Segmentation by Predicting Signed Distance Maps
Yuan Xue, Hui Tang, Zhi Qiao, Guanzhong Gong, Yong Yin, Zhen Qian, Chao Huang, Wei Fan, Xiaolei Huang
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 joint SDM prediction and Dice segmentation training | Deep 3D UNet on the hippocampus test set · Hippocampus | HD (mm): 23.568\pm17.811 -> 5.400\pm1.895 | Improved |
| Add joint SDM prediction and Dice segmentation training | Deep 3D UNet on the hippocampus test set · Hippocampus | Dice: 0.840\pm0.025 -> 0.843\pm0.032 | No clear effect |
| Add joint SDM prediction and Dice segmentation training | Deep 3D UNet on the hippocampus test set · Hippocampus | ASD (mm): 0.414\pm0.153 -> 0.345\pm0.130 | Improved |
| Use SDM-only training instead of joint SDM and segmentation-map training (ablation) | Deep 3D UNet on the hippocampus test set · Hippocampus | Dice: 0.843\pm0.032 -> 0.757\pm0.065 | Hurt |
| Add joint SDM prediction and Dice segmentation training | Deep 3D UNet on MICCAI 2015 · MICCAI 2015 | Average HD95 (mm): 3.32 -> 1.98 | Improved |
| Add joint SDM prediction and Dice segmentation training | Deep 3D UNet on MICCAI 2015 · MICCAI 2015 | Average Dice: 0.859 -> 0.845 | Hurt |
| Add joint SDM prediction and Dice segmentation training | Deep 3D UNet on MICCAI 2015 · MICCAI 2015 | Average ASD (mm): 0.44 -> 0.39 | Improved |
| Replace joint SDM + Dice training with SDM-only training (ablation) | Deep 3D UNet on MICCAI 2015 · MICCAI 2015 | Average ASD (mm): 0.39 -> 1.55 | Hurt |
| Replace the proposed SDM regression loss with L1 loss (ablation) | Joint SDM and Dice training on MICCAI 2015 · MICCAI 2015 | Average HD (mm): 5.07 -> 7.47 | Hurt |
| Replace the proposed SDM regression loss with L1 loss (ablation) | Joint SDM and Dice training on MICCAI 2015 · MICCAI 2015 | Average ASD (mm): 0.39 -> 0.54 | Hurt |
| Use a Deep 3D UNet with six downsampling operations rather than one (ablation) | Backbone segmentation network on MICCAI 2015 · MICCAI 2015 | Segmentation performance | Improved |
Closest papers by meaning
- DoDNet: Learning To Segment Multi-Organ and Tumors From Multiple Partially Labeled Datasets (CVPR 2021 · similarity 0.86)
- Continual Segment: Towards a Single, Unified and Non-forgetting Continual Segmentation Model of 143 Whole-body Organs in CT Scans (ICCV 2023 · similarity 0.86)
- Teaching AI the Anatomy Behind the Scan: Addressing Anatomical Flaws in Medical Image Segmentation with Learnable Prior (ICCV 2025 · similarity 0.85)
- Structure Boundary Preserving Segmentation for Medical Image With Ambiguous Boundary (CVPR 2020 · similarity 0.85)
- Deep Implicit Statistical Shape Models for 3D Medical Image Delineation (AAAI 2022 · similarity 0.85)
- Towards a Comprehensive, Efficient and Promptable Anatomic Structure Segmentation Model Using 3D Whole-Body CT Scans (AAAI 2025 · 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.
