Research map / Continual learning / Online and replay-based
Distributed Weight Consolidation: A Brain Segmentation Case Study
Patrick McClure, Charles Y Zheng, Jakub Kaczmarzyk, John Rogers-Lee, Satra Ghosh, Dylan Nielson, Peter A Bandettini, Francisco Pereira
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 |
|---|---|---|---|
| Consolidate distributed VCL networks and fine-tune on H | MeshNet DWC · HCP, NKI, Buckner, WU120 weighted average | Dice: 75.94 -> 76.82 | Improved |
| Consolidate distributed VCL networks and fine-tune on H | MeshNet DWC · ABIDE | Dice: 66.27 -> 66.21; p = 0.733 | No clear effect |
| Omit fine-tuning on H after consolidation (ablation) | MeshNet DWC · HCP, NKI, Buckner, WU120 weighted average | Dice: 75.95 -> 76.82 | Hurt |
| Omit fine-tuning on H after consolidation (ablation) | MeshNet DWC · ABIDE | Dice: 65.56 -> 66.21 | Hurt |
| Use increasing-size sequential VCL order H → N → B → W (ablation) | MeshNet continual learning · HCP, NKI, Buckner, WU120 weighted average | Dice: 74.86 vs 76.82 | Hurt |
| Use increasing-size sequential VCL order H → N → B → W (ablation) | MeshNet continual learning · ABIDE | Dice: 63.3 vs 66.21 | Hurt |
| Use decreasing-size sequential VCL order H → W → B → N (ablation) | MeshNet continual learning · HCP, NKI, Buckner, WU120 weighted average | Dice: 77.99 vs 76.82 | Mixed |
| Use decreasing-size sequential VCL order H → W → B → N (ablation) | MeshNet continual learning · ABIDE | Dice: 70.79 vs 66.21 | Mixed |
| Change dataset order in sequential VCL (ablation) | MeshNet continual learning · HCP, NKI, Buckner, WU120 | Dice: 74.86 vs 77.99 | Mixed |
Closest papers by meaning
- Similarity Memory Prior is All You Need for Medical Image Segmentation (ICCV 2025 · similarity 0.83)
- Breaking Task Boundaries: A Unified Model for 3D Medical Image Fusion and Segmentation Guided by Manifold Perspective (AAAI 2026 · similarity 0.83)
- Continual Segment: Towards a Single, Unified and Non-forgetting Continual Segmentation Model of 143 Whole-body Organs in CT Scans (ICCV 2023 · similarity 0.83)
- DualNet: Continual Learning, Fast and Slow (NeurIPS 2021 · similarity 0.83)
- FR²Seg: Continual Segmentation Across Multiple Sites via Fourier Style Replay and Adaptive Consistency Regularization (AAAI 2025 · similarity 0.82)
- Dynamic Neural Regeneration: Enhancing Deep Learning Generalization on Small Datasets (NeurIPS 2024 · 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.
