PADReg: Physics-Aware Deformable Registration Guided by Contact Force for Ultrasound Sequences

2025-08-12

AI Summary

Ultrasound elastography registration under large deformations suffers from poor anatomical consistency and limited biomechanical interpretability due to low image contrast, severe noise, and ill-defined tissue boundaries. To address this, we propose a deformable registration framework incorporating contact-force–guided physical priors: for the first time, we jointly leverage robotically acquired contact forces and pixel-wise stiffness maps to construct a lightweight, Hooke’s law–constrained physics-aware module that indirectly estimates dense deformation fields from multimodal ultrasound images. Our method significantly improves both anatomical alignment accuracy and biomechanical plausibility. Evaluated on an in vivo dataset, it achieves an HD95 of 12.90 mm—21.34% lower than the state-of-the-art—demonstrating its potential for high-precision, interpretable registration in thyroid and breast disease diagnosis.

Technology Category

• Computer Vision: Low Level & Physics-based Vision
• Intelligent Robots: Manipulation
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Ultrasound deformable registration estimates spatial transformations between pairs of deformed ultrasound images, which is crucial for capturing biomechanical properties and enhancing diagnostic accuracy in diseases such as thyroid nodules and breast cancer. However, ultrasound deformable registration remains highly challenging, especially under large deformation. The inherently low contrast, heavy noise and ambiguous tissue boundaries in ultrasound images severely hinder reliable feature extraction and correspondence matching. Existing methods often suffer from poor anatomical alignment and lack physical interpretability. To address the problem, we propose PADReg, a physics-aware deformable registration framework guided by contact force. PADReg leverages synchronized contact force measured by robotic ultrasound systems as a physical prior to constrain the registration. Specifically, instead of directly predicting deformation fields, we first construct a pixel-wise stiffness map utilizing the multi-modal information from contact force and ultrasound images. The stiffness map is then combined with force data to estimate a dense deformation field, through a lightweight physics-aware module inspired by Hooke's law. This design enables PADReg to achieve physically plausible registration with better anatomical alignment than previous methods relying solely on image similarity. Experiments on in-vivo datasets demonstrate that it attains a HD95 of 12.90, which is 21.34% better than state-of-the-art methods. The source code is available at https://github.com/evelynskip/PADReg.

Problem

• Estimates spatial transformations in deformed ultrasound images
• Addresses poor alignment and physical interpretability in registration
• Improves registration accuracy using contact force and stiffness maps

Innovation

• Physics-aware deformable registration guided by contact force
• Pixel-wise stiffness map from force and ultrasound data
• Lightweight physics-aware module inspired by Hooke's law