3D Bone Model Generation from 2D X-ray Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current patient-specific total joint replacement systems rely heavily on MRI scans for accurate joint anatomy interpretation, which is costly and not accessible in all regions, leading to inefficiencies and potential delays in surgical procedures.
Innovation Solution
A computer-assisted surgical system that utilizes bi-planar two-dimensional images to create three-dimensional anatomical models, reducing the reliance on expensive MRI scans and enabling more accessible and efficient surgical planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If MRI scans are used for patient-specific joint replacement planning, then measurement precision and manufacturing precision are improved, but cost and accessibility worsen
Solution Approach 1:
The patent uses 2D X-ray images as simplified copies or representations of the bone anatomy, replacing the need for expensive 3D MRI scans. The system creates a 3D bone model by processing multiple 2D X-ray images, effectively using a simpler imaging modality to achieve the necessary anatomical information for surgical planning.
Solution Approach 2:
The patent replaces the complex mathematical calculations and processing systems traditionally required to convert 2D images to 3D models with a machine learning-based system. The deep learning model automatically processes 2D X-ray images to generate accurate 3D bone models, eliminating the need for complex manual or algorithmic transformations.
2Manufacturing precision
If complex mathematical calculations are used to convert 2D images to 3D models, then manufacturing precision is improved, but device complexity and computational requirements worsen
Solution Approach 1:
The patent replaces complex mathematical calculations and traditional image processing algorithms with a machine learning-based system. A deep learning model is trained to directly convert 2D X-ray images into accurate 3D bone models, eliminating the need for complex iterative optimization algorithms and reducing computational complexity significantly.
Solution Approach 2:
The patent changes the approach from using complex mathematical parameters and iterative optimization to using a pre-trained neural network model. The system transforms the problem from one requiring complex real-time calculations to one involving straightforward data input and model inference, dramatically reducing computational requirements.
3Measurement precision
If mathematical optimization algorithms are used to match bone shapes, then measurement precision is improved, but loss of time and productivity worsen
Solution Approach 1:
The patent replaces time-consuming mathematical optimization algorithms with a machine learning-based system. The deep learning model is pre-trained on extensive datasets to perform bone shape matching and 3D model generation, eliminating the need for iterative optimization processes and significantly reducing processing time.
Solution Approach 2:
The patent applies preliminary action by pre-training the machine learning model on a comprehensive dataset of bone images and corresponding 3D models. This pre-training phase captures the essential patterns and relationships needed for accurate bone shape matching, allowing the system to quickly generate accurate 3D models from new 2D X-ray images without requiring time-consuming optimization.
Data Source
AI summary
A method of generating a custom three-dimensional (3D) model of a patient bone from one or more 2D images is disclosed. The method includes obtaining a 2D image of a bone, optionally of a joint, and identifying a 3D bone template for a candidate or representative bone from a pre-aligned library of representative bones. The method further includes repositioning one or more views of the 3D model or 2D images (e.g., with respect to rotation angle or caudal angle). In an iterative process, another 3D bone model for another candidate bone can be identified based on the repositioning until an accuracy threshold is satisfied. When the accuracy threshold is satisfied, surface region(s) of the current 3D bone model can then be modified to generate the resulting 3D model for the patient bone. The process can then be repeated for other bone(s) associated with the joint of the patient.