Surgical Robot Image Registration Using Guided 2D/3D Pose Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 2D/3D image registration methods for surgical robots lack accurate and efficient calculation of initial pose and similarity between CT and X-ray images, leading to low registration accuracy and efficiency.
Innovation Solution
An image registration method that captures 2D and 3D images, calculates pose information using guiding information, adjusts 3D images, and uses normalized cross-correlation coefficients to determine similarity, iteratively updating pose information until registration criteria are met.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional 2D/3D image registration methods are used, then the registration process can be completed, but the registration accuracy is low and the efficiency is slow
Solution Approach 1:
The patent applies preliminary action by pre-calculating pose information from guiding information in both 2D and 3D images before the actual registration process. This preliminary pose estimation provides a good initial state for the registration algorithm, enabling faster convergence and higher accuracy without sacrificing time. The guiding information extraction and initial pose calculation are performed in advance to prepare optimal starting conditions for the registration task.
Solution Approach 2:
The patent replaces traditional mechanical/image-based registration methods with a computational approach using normalized cross-correlation coefficients and pose information calculation. Instead of relying on manual alignment or simple image matching, the system uses algorithmic pose estimation from guiding information and mathematical optimization through similarity metrics, achieving both higher accuracy and efficiency.
2Measurement precision
If the similarity calculation is performed iteratively with pose updates, then the registration accuracy improves, but the computational time increases
Solution Approach 1:
The patent implements feedback by iteratively updating pose information based on similarity calculations between 2D images and digitally reconstructed radiographs. The normalized cross-correlation coefficient provides a quantitative feedback metric that guides the optimization process. When the similarity meets the preset condition, the iteration stops, ensuring both high accuracy and reasonable computational time through adaptive feedback control.
Solution Approach 2:
The patent applies parameter changes by dynamically adjusting pose parameters (position and orientation) based on the similarity metric. The system changes the pose parameters iteratively to maximize the normalized cross-correlation coefficient, achieving accurate registration. The preset condition on similarity acts as a threshold parameter that controls when to stop the parameter optimization, balancing accuracy and time.
Data Source
AI summary
An image registration method for a surgical robot, comprising: capturing a two-dimensional image of a surgical object and determining first guiding information in the two-dimensional image; acquiring a three-dimensional image of the surgical object and preoperative planning information, and determining second guiding information in the three-dimensional image; calculating first pose information according to the first guiding information and the second guiding information; adjusting the three-dimensional image according to the first pose information, and acquiring a digitally reconstructed two-dimensional image in the adjusted three-dimensional image; calculating the similarity between the two-dimensional image and the digitally reconstructed two-dimensional image; when the similarity meets a preset condition, projecting the preoperative planning information onto the two-dimensional image according to the first pose information; when the similarity does not meet the preset condition, updating the first pose information to obtain second pose information, and adjusting the three-dimensional image according to the second pose information.