Context-Gated Cross-Modal Perception with Visual Mamba for PET-CT Lung Tumor Segmentation

2025-10-31

AI Summary

To address challenges in PET-CT multimodal lung tumor segmentation—including difficult anatomical–functional information fusion and high computational overhead—this paper proposes vMambaX, a lightweight framework. Its core is the Context-Gated Cross-modal Perception (CGCP) module, which employs a dynamic gating mechanism to adaptively emphasize discriminative regions while suppressing modality-specific noise, thereby enabling efficient, accurate, and multi-scale interaction between PET and CT features. Built upon the Visual Mamba architecture, vMambaX significantly reduces parameter count and computational complexity. Evaluated on the PCLT20K dataset, vMambaX achieves superior segmentation accuracy (e.g., +2.3% Dice score) with substantially lower resource consumption, demonstrating its effectiveness, efficiency, and scalability for precision diagnosis and treatment of lung cancer.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Multimodal Learning
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

Accurate lung tumor segmentation is vital for improving diagnosis and treatment planning, and effectively combining anatomical and functional information from PET and CT remains a major challenge. In this study, we propose vMambaX, a lightweight multimodal framework integrating PET and CT scan images through a Context-Gated Cross-Modal Perception Module (CGM). Built on the Visual Mamba architecture, vMambaX adaptively enhances inter-modality feature interaction, emphasizing informative regions while suppressing noise. Evaluated on the PCLT20K dataset, the model outperforms baseline models while maintaining lower computational complexity. These results highlight the effectiveness of adaptive cross-modal gating for multimodal tumor segmentation and demonstrate the potential of vMambaX as an efficient and scalable framework for advanced lung cancer analysis. The code is available at https://github.com/arco-group/vMambaX.

Problem

• Segmenting lung tumors from PET-CT scans accurately
• Enhancing multimodal feature interaction while suppressing noise
• Developing lightweight efficient framework for cancer analysis

Innovation

• Integrates PET-CT scans via context-gated cross-modal perception
• Uses Visual Mamba architecture for adaptive feature enhancement
• Lightweight framework reduces noise while maintaining computational efficiency