CenterMamba-SAM: Center-Prioritized Scanning and Temporal Prototypes for Brain Lesion Segmentation
Not on the map. Why?
Abstract (arXiv)
Brain lesion segmentation remains challenging due to small, low-contrast lesions, anisotropic sampling, and cross-slice discontinuities. We propose CenterMamba-SAM, an end-to-end framework that freezes a pretrained backbone and trains only lightweight adapters for efficient fine-tuning. At its core is the CenterMamba encoder, which employs a novel 3x3 corner-axis-center short-sequence scanning strategy to enable center-prioritized, axis-reinforced, and diagonally compensated information aggregation. This design enhances sensitivity to weak boundaries and tiny foci while maintaining sparse yet effective feature representation. A memory-driven structural prompt generator maintains a prototype bank across neighboring slices, enabling automatic synthesis of reliable prompts without user interaction, thereby improving inter-slice coherence. The memory-augmented multi-scale decoder integrates memory attention modules at multiple levels, combining deep supervision with progressive refinement to restore fine details while preserving global consistency. Extensive experiments on public benchmarks demonstrate that CenterMamba-SAM achieves state-of-the-art performance in brain lesion segmentation.
Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 23 of its 33 references that are in the map. First posted on arXiv 3 Nov 2025.
Builds on
23 of 33 references.
References · 23
- 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
- 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,082
- 2023Segment Anythingcited by 8,158
- 2019EfficientNet: Rethinking Model Scaling for Convolutional Neural Networkscited by 6,415
- 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
- 2020Training data-efficient image transformers & distillation through attentioncited by 4,560
- 2024SAM 2: Segment Anything in Images and Videoscited by 2,784
- 2021TransUNet: Transformers Make Strong Encoders for Medical Image Segmentationcited by 1,552
- 2021Swin-Unet: Unet-like Pure Transformer for Medical Image Segmentationcited by 1,218
- 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
- 2024VMamba: Visual State Space Modelcited by 955
- 2020Designing Network Design Spacescited by 954
11 more
- 2021UNETR: Transformers for 3D Medical Image Segmentationcited by 954
- 2017Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentationscited by 614
- 2021The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classificationcited by 507
- 2023Segment Anything in Medical Imagescited by 226
- 2024EfficientVMamba: Atrous Selective Scan for Light Weight Visual Mambacited by 146
- 2022ISLES 2022: A multi-center magnetic resonance imaging stroke lesion segmentation datasetcited by 104
- 2023STU-Net: Scalable and Transferable Medical Image Segmentation Models Empowered by Large-Scale Supervised Pre-trainingcited by 96
- 2024RainMamba: Enhanced Locality Learning with State Space Models for Video Derainingcited by 56
- 2023The state-of-the-art 3D anisotropic intracranial hemorrhage segmentation on non-contrast head CT: The INSTANCE challengecited by 25
- 2025SETransformer: A Hybrid Attention-Based Architecture for Robust Human Activity Recognitioncited by 19
- 2025Robust Anomaly Detection in Network Traffic: Evaluating Machine Learning Models on CICIDS2017cited by 4
Built on by
Citing papers per year: 2025: 1; 2026: 7.
8 papers.
- 2025Adaptive Detector-Verifier Framework for Zero-Shot Polyp Detection in Open-World Settingscited by 4
- 2026Adaptive Multi-Stage Patent Claim Generation with Unified Quality Assessmentcited by 0
- 2026EVLA: An Electro-Aware Multimodal Assistant for Physically-Grounded Driving Reasoning and Controlcited by 0
- 2026A3M: Adaptive, Adversarial and Multi-Objective Learning for Strategic Bidding in Repeated Auctionscited by 0
- 2026Beyond Skepticism: Evaluating LLMs Pedagogical Intent Reasoning with the Adaptive Pedagogical Vigilance Frameworkcited by 0
- 2026FedCausal-Dyn: A Causal-Dynamic Paradigm for Federated Learning under Dynamic Feature Driftcited by 0
- 2026Graph-Based Modeling of Financial Volatility Dynamicscited by 0
- 2026Style Over Substance: Content-Invariant Wrappers Flip LLM Safety-Judge Verdictscited by 0
- 2026Style Over Substance: Content-Invariant Wrappers Flip LLM Safety-Judge Verdicts8 Sep 2026
- 2026Beyond Skepticism: Evaluating LLMs Pedagogical Intent Reasoning with the Adaptive Pedagogical Vigilance Framework2 Jul 2026
- 2026A3M: Adaptive, Adversarial and Multi-Objective Learning for Strategic Bidding in Repeated Auctions27 Jun 2026
- 2026EVLA: An Electro-Aware Multimodal Assistant for Physically-Grounded Driving Reasoning and Control27 Jun 2026
- 2026Graph-Based Modeling of Financial Volatility Dynamics23 Jun 2026
- 2026FedCausal-Dyn: A Causal-Dynamic Paradigm for Federated Learning under Dynamic Feature Drift20 Jun 2026
- 2026Adaptive Multi-Stage Patent Claim Generation with Unified Quality Assessment14 Jan 2026
- 2025Adaptive Detector-Verifier Framework for Zero-Shot Polyp Detection in Open-World Settings13 Dec 2025