Weight-Bearing Pose Estimation for 3D Bone Models in Surgical Planning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current total ankle replacement (TAR) surgical planning methods fail to accurately account for weight-bearing conditions, as 3D bone models generated from non-weight bearing images do not reflect the actual weighted position of bones, leading to less than desirable surgical outcomes.
Innovation Solution
A method that maps weight-bearing conditions from standing X-ray images to 3D bone models generated from non-weight bearing medical images, involving surface projection, contour extraction, shape matching, and pose updating to align the bone models with weighted poses, using both manual and automated processes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If 3D bone models are generated from non-weight bearing medical images, then the imaging process is simpler and faster, but the accuracy of surgical planning is reduced
Solution Approach 1:
The patent introduces X-ray images as an intermediary element that captures weight-bearing bone positions. These X-ray images serve as a mediator between the non-weight-bearing CT scans and the surgical planning requirements, allowing the system to infer weighted bone poses without requiring weighted CT scans. The X-ray images bridge the gap between available imaging data and surgical planning needs.
Solution Approach 2:
The patent creates 2D projections (copies) of the 3D bone models and compares them with 2D X-ray images. By working with 2D projections rather than directly manipulating 3D weighted scans, the system can leverage existing non-weight-bearing CT data while accounting for weight-bearing conditions through comparison with X-ray copies. This copying approach allows indirect representation of weighted bone positions.
2Device complexity
If traditional surgical planning methods are used without weight-bearing considerations, then the planning process is simpler, but the surgical outcomes are less desirable
Solution Approach 1:
The patent transitions from 3D bone models to 2D projections for comparison with X-ray images. This dimensionality reduction simplifies the matching process while still capturing essential pose information. By working in 2D space for the critical comparison step, the system reduces computational complexity while maintaining accuracy in determining weight-bearing bone positions.
Solution Approach 2:
The patent implements an iterative registration process where 2D projections of the 3D bone model are compared with X-ray images, and the model pose is updated based on the comparison. This feedback loop continues until convergence, allowing the system to progressively refine the bone pose estimation to accurately reflect weight-bearing conditions while maintaining a relatively simple overall process.
Data Source
AI summary
A system for processing patient data associated with a joint of a patient. The system including a computing device executing instructions to provide a bone model of the joint that is representative of the joint of the patient in a first pose. The instructions further including rearranging the bone models relative to each other to mimic a second pose of the joint of the patient from at least one image or model representative of the joint of the patient in the second pose. The instructions further including planning a joint replacement procedure with the bone models after rearranging the plurality of bone models to mimic the second pose of the joint of the patient from the at least one image or model representative of the joint of the patient in the second pose.