Dynamic Stream Network for Combinatorial Explosion Problem in Deformable Medical Image Registration

2025-12-22

AI Summary

In deformable medical image registration (DMIR), dual-input architectures suffer from feature combination explosion—redundant and weakly correlated feature interactions hinder high-fidelity deformation modeling. To address this, we propose AdSB-DySA: (1) an Adaptive Stream Basin (AdSB) module that dynamically modulates receptive fields to suppress irrelevant spatial responses; and (2) a Dynamic Flow Self-Attention (DySA) mechanism that generates sample- and channel-wise dynamic weights to precisely capture cross-image, highly correlated latent feature relationships. The framework jointly integrates dynamic receptive field modeling, dynamic weight generation, and multi-scale feature disentanglement-recomposition. Evaluated on standard benchmarks—including LPBA40, OASIS, and HIRA—AdSB-DySA consistently outperforms state-of-the-art methods, achieving significant improvements in registration accuracy (e.g., Dice score, Jacobian determinant ratio) and cross-domain generalization capability.

Technology Category

• Computer Vision: Motion & Tracking
• Search and Optimization: Sampling/Simulation-based Search
• Multiagent Systems: Adversarial Agents

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Vertical and domain-specific search
• Systems and Infrastructure for Web, Mobile and WoT: Data management and stream processing for Web, mobile and wireless applications

Abstract

Combinatorial explosion problem caused by dual inputs presents a critical challenge in Deformable Medical Image Registration (DMIR). Since DMIR processes two images simultaneously as input, the combination relationships between features has grown exponentially, ultimately the model considers more interfering features during the feature modeling process. Introducing dynamics in the receptive fields and weights of the network enable the model to eliminate the interfering features combination and model the potential feature combination relationships. In this paper, we propose the Dynamic Stream Network (DySNet), which enables the receptive fields and weights to be dynamically adjusted. This ultimately enables the model to ignore interfering feature combinations and model the potential feature relationships. With two key innovations: 1) Adaptive Stream Basin (AdSB) module dynamically adjusts the shape of the receptive field, thereby enabling the model to focus on the feature relationships with greater correlation. 2) Dynamic Stream Attention (DySA) mechanism generates dynamic weights to search for more valuable feature relationships. Extensive experiments have shown that DySNet consistently outperforms the most advanced DMIR methods, highlighting its outstanding generalization ability. Our code will be released on the website: https://github.com/ShaochenBi/DySNet.

Problem

• Addresses combinatorial explosion in deformable medical image registration
• Reduces interfering feature combinations via dynamic receptive fields
• Models potential feature relationships with adaptive weights and attention

Innovation

• Dynamic Stream Network with adaptive receptive fields
• Adaptive Stream Basin module for correlation focus
• Dynamic Stream Attention mechanism for valuable features