TCIP: Threshold-Controlled Iterative Pyramid Network for Deformable Medical Image Registration

2025-10-08

AI Summary

In deformable medical image registration, pyramidal network decoders often accumulate anatomical misalignments, and fixed iteration counts hinder accurate convergence. To address these issues, this paper proposes a Threshold-Controlled Iterative Pyramidal network (TCIP). Methodologically: (1) a Feature-Enhanced Residual Module (FERM) is designed to suppress irrelevant feature propagation and mitigate misalignment accumulation; (2) a two-stage Threshold-Controlled Iteration (TCI) strategy is introduced, dynamically terminating iterations based on the stability and convergence of the deformation field. TCIP enables end-to-end registration and outperforms state-of-the-art methods on three brain MRI and one abdominal CT datasets, achieving higher registration accuracy, faster inference speed, and fewer parameters. It further demonstrates strong generalizability and ablation-validated module efficacy.

Technology Category

• Computer Vision: Multi-modal Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Although pyramid networks have demonstrated superior performance in deformable medical image registration, their decoder architectures are inherently prone to propagating and accumulating anatomical structure misalignments. Moreover, most existing models do not adaptively determine the number of iterations for optimization under varying deformation requirements across images, resulting in either premature termination or excessive iterations that degrades registration accuracy. To effectively mitigate the accumulation of anatomical misalignments, we propose the Feature-Enhanced Residual Module (FERM) as the core component of each decoding layer in the pyramid network. FERM comprises three sequential blocks that extract anatomical semantic features, learn to suppress irrelevant features, and estimate the final deformation field, respectively. To adaptively determine the number of iterations for varying images, we propose the dual-stage Threshold-Controlled Iterative (TCI) strategy. In the first stage, TCI assesses registration stability and with asserted stability, it continues with the second stage to evaluate convergence. We coin the model that integrates FERM and TCI as Threshold-Controlled Iterative Pyramid (TCIP). Extensive experiments on three public brain MRI datasets and one abdomen CT dataset demonstrate that TCIP outperforms the state-of-the-art (SOTA) registration networks in terms of accuracy, while maintaining comparable inference speed and a compact model parameter size. Finally, we assess the generalizability of FERM and TCI by integrating them with existing registration networks and further conduct ablation studies to validate the effectiveness of these two proposed methods.

Problem

• Pyramid networks propagate anatomical misalignment errors in registration
• Existing models lack adaptive iteration control for varying deformation needs
• Current methods suffer from premature termination or excessive iterations

Innovation

• Feature-Enhanced Residual Module reduces anatomical misalignment propagation
• Threshold-Controlled Iterative strategy adaptively determines iteration numbers
• Pyramid network integrates dual-stage convergence assessment for deformable registration