Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

CellMamba: Adaptive Mamba for Accurate and Efficient Cell Detection

Ruochen Liu, Yi Tian, Jiahao Wang, Hongbin Liu, Xianxu Hou, Jingxin Liu

TL;DR

CellMamba targets accurate, efficient nucleus/cell detection in high-resolution pathology images by marrying a mixed Mamba-Transformer backbone with a novel Triple-Mapping Adaptive Coupling (TMAC) module and an Adaptive Mamba Head. TMAC employs dual-channel splitting and three attention maps to refine local spatial cues and suppress background noise, while the adaptive head fuses multi-scale features with learnable weights for varying cell sizes. On CoNSeP and CytoDArk0, CellMamba achieves state-of-the-art mAP with a compact 14.7M-parameter footprint and 1.6 ms inference per 256×256 patch, outperforming CNN-, Transformer-, and other Mamba-based baselines. This work demonstrates that structured state-space approaches can deliver both high accuracy and practical efficiency for dense, fine-grained biomedical detection tasks.

Abstract

Cell detection in pathological images presents unique challenges due to densely packed objects, subtle inter-class differences, and severe background clutter. In this paper, we propose CellMamba, a lightweight and accurate one-stage detector tailored for fine-grained biomedical instance detection. Built upon a VSSD backbone, CellMamba integrates CellMamba Blocks, which couple either NC-Mamba or Multi-Head Self-Attention (MSA) with a novel Triple-Mapping Adaptive Coupling (TMAC) module. TMAC enhances spatial discriminability by splitting channels into two parallel branches, equipped with dual idiosyncratic and one consensus attention map, adaptively fused to preserve local sensitivity and global consistency. Furthermore, we design an Adaptive Mamba Head that fuses multi-scale features via learnable weights for robust detection under varying object sizes. Extensive experiments on two public datasets-CoNSeP and CytoDArk0-demonstrate that CellMamba outperforms both CNN-based, Transformer-based, and Mamba-based baselines in accuracy, while significantly reducing model size and inference latency. Our results validate CellMamba as an efficient and effective solution for high-resolution cell detection.

CellMamba: Adaptive Mamba for Accurate and Efficient Cell Detection

TL;DR

CellMamba targets accurate, efficient nucleus/cell detection in high-resolution pathology images by marrying a mixed Mamba-Transformer backbone with a novel Triple-Mapping Adaptive Coupling (TMAC) module and an Adaptive Mamba Head. TMAC employs dual-channel splitting and three attention maps to refine local spatial cues and suppress background noise, while the adaptive head fuses multi-scale features with learnable weights for varying cell sizes. On CoNSeP and CytoDArk0, CellMamba achieves state-of-the-art mAP with a compact 14.7M-parameter footprint and 1.6 ms inference per 256×256 patch, outperforming CNN-, Transformer-, and other Mamba-based baselines. This work demonstrates that structured state-space approaches can deliver both high accuracy and practical efficiency for dense, fine-grained biomedical detection tasks.

Abstract

Cell detection in pathological images presents unique challenges due to densely packed objects, subtle inter-class differences, and severe background clutter. In this paper, we propose CellMamba, a lightweight and accurate one-stage detector tailored for fine-grained biomedical instance detection. Built upon a VSSD backbone, CellMamba integrates CellMamba Blocks, which couple either NC-Mamba or Multi-Head Self-Attention (MSA) with a novel Triple-Mapping Adaptive Coupling (TMAC) module. TMAC enhances spatial discriminability by splitting channels into two parallel branches, equipped with dual idiosyncratic and one consensus attention map, adaptively fused to preserve local sensitivity and global consistency. Furthermore, we design an Adaptive Mamba Head that fuses multi-scale features via learnable weights for robust detection under varying object sizes. Extensive experiments on two public datasets-CoNSeP and CytoDArk0-demonstrate that CellMamba outperforms both CNN-based, Transformer-based, and Mamba-based baselines in accuracy, while significantly reducing model size and inference latency. Our results validate CellMamba as an efficient and effective solution for high-resolution cell detection.
Paper Structure (18 sections, 10 equations, 2 figures, 4 tables)

This paper contains 18 sections, 10 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Backbone Architecture
  5. Triple-Mapping Adaptive Coupling
  6. Channel Splitting
  7. Triple Mapping
  8. Adaptive Coupling
  9. Feature Fusion
  10. Adaptive Mamba Head
  11. Experiments
  12. Datasets
  13. Implementation Details
  14. Experimental Results
  15. Comparison with State-of-the-art.
  16. ...and 3 more sections

Figures (2)

  • Figure 1: The overall framework architecture of CellMamba, comprises a four-stage mixed Mamba-Transformer hierarchical backbone and an adaptive Mamba head for cell classification and box regression.
  • Figure 2: Detailed structure of the proposed CellMamba block. TMAC is placed after sequence attention and before Feed-Forward Network (FFN), enabling local spatial refinement via triple mapping and adaptive coupling.