# MambaBack: Bridging Local Features and Global Contexts in Whole Slide Image Analysis

arXiv: https://arxiv.org/abs/2604.15729 · HTML: https://mlsanity.com/paper/2604.15729/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Mamba for medical image analysis, one of 10 directions in State-space sequence models (S4, Mamba).
- Builds on 14 of its 31 references that are in the map.
- First posted on arXiv 17 Apr 2026.

## Placed in

- State-space sequence models (S4, Mamba) / [Mamba for medical image analysis](https://mlsanity.com/ssm/mamba-for-medical-image-analysis/)

## Builds on (most cited first, 14 of 14)

- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality](https://mlsanity.com/paper/2405.21060/) (2024)
- [A Survey on Deep Learning in Medical Image Analysis](https://mlsanity.com/paper/1702.05747/) (2017)
- [Attention-based Deep Multiple Instance Learning](https://mlsanity.com/paper/1802.04712/) (2018)
- [TransMIL: Transformer based Correlated Multiple Instance Learning for Whole Slide Image Classification](https://mlsanity.com/paper/2106.00908/) (2021)
- [DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network](https://mlsanity.com/paper/1606.00931/) (2016)
- [DTFD-MIL: Double-Tier Feature Distillation Multiple Instance Learning for Histopathology Whole Slide Image Classification](https://mlsanity.com/paper/2203.12081/) (2022)
- [Deep neural network models for computational histopathology: A survey](https://mlsanity.com/paper/1912.12378/) (2019)
- [MambaOut: Do We Really Need Mamba for Vision?](https://mlsanity.com/paper/2405.07992/) (2024)
- [MambaMIL: Enhancing Long Sequence Modeling with Sequence Reordering in Computational Pathology](https://mlsanity.com/paper/2403.06800/) (2024)
- [A Survey of FPGA-Based Neural Network Accelerator](https://mlsanity.com/paper/1712.08934/) (2017)
- [UnPuzzle: A Unified Framework for Pathology Image Analysis](https://mlsanity.com/paper/2503.03152/) (2025)
- [CellMix: A General Instance Relationship based Method for Data Augmentation Towards Pathology Image Classification](https://mlsanity.com/paper/2301.11513/) (2023)
- [PathRWKV: Enhancing Whole Slide Image Inference with Asymmetric Recurrent Modeling](https://mlsanity.com/paper/2503.03199/) (2025)

## Built on by (most cited first, 1 of 1)

- [A Multi-Cohort Validation of Censoring-Aware Conformal Lower Predictive Bounds for Pathology Survival Models](https://mlsanity.com/paper/2608.04025/) (2026)

## Abstract (arXiv)

Whole Slide Image (WSI) analysis is pivotal in computational pathology, enabling cancer diagnosis by integrating morphological and architectural cues across magnifications. Multiple Instance Learning (MIL) serves as the standard framework for WSI analysis. Recently, Mamba has become a promising backbone for MIL, overtaking Transformers due to its efficiency and global context modeling capabilities originating from Natural Language Processing (NLP). However, existing Mamba-based MIL approaches face three critical challenges: (1) disruption of 2D spatial locality during 1D sequence flattening; (2) sub-optimal modeling of fine-grained local cellular structures; and (3) high memory peaks during inference on resource-constrained edge devices. Studies like MambaOut reveal that Mamba's SSM component is redundant for local feature extraction, where Gated CNNs suffice. Recognizing that WSI analysis demands both fine-grained local feature extraction akin to natural images, and global context modeling akin to NLP, we propose MambaBack, a novel hybrid architecture that harmonizes the strengths of Mamba and MambaOut. First, we propose the Hilbert sampling strategy to preserve the 2D spatial locality of tiles within 1D sequences, enhancing the model's spatial perception. Second, we design a hierarchical structure comprising a 1D Gated CNN block based on MambaOut to capture local cellular features, and a BiMamba2 block to aggregate global context, jointly enhancing multi-scale representation. Finally, we implement an asymmetric chunking design, allowing parallel processing during training and chunking-streaming accumulation during inference, minimizing peak memory usage for deployment. Experimental results on five datasets demonstrate that MambaBack outperforms seven state-of-the-art methods. Source code and datasets are publicly available.