3D Joint Model from 2D Images for Surgical Precision
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current orthopedic surgical techniques face challenges in accurately visualizing and removing bone during minimally-invasive procedures due to limited field of view and distortion in endoscopic images, as well as the limitations of X-ray imaging in providing a comprehensive view of joint pathology.
Innovation Solution
A method and system that generate a three-dimensional model of a joint using machine learning models trained on imaging data from two-dimensional imaging modalities, allowing for the creation of a detailed, accurate representation of the joint's state during a medical procedure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If endoscopic imaging is used to view the debridement area, then the surgeon can observe the surgical site, but the field of view is limited and the image is distorted making it difficult to determine the extent of bone removal
Solution Approach 1:
The patent transforms two-dimensional endoscopic images into a three-dimensional model of the joint, allowing the surgeon to view the debridement area from multiple angles and perspectives. This dimensional transformation provides a comprehensive view of the entire pathology and the extent of bone removal, overcoming the limited field of view and distortion inherent in traditional endoscopic imaging.
2Measurement precision
If X-ray imaging is used to observe the bone perimeter, then the surgeon can assess where and how much bone should be removed, but only the horizon of the bone is observable and it is difficult to compare with arthroscopic imaging
Solution Approach 1:
The patent merges arthroscopic imaging data with X-ray imaging data into a single integrated three-dimensional model. This combination allows the surgeon to simultaneously observe the bone perimeter measurements from X-ray imaging and the visual appearance of the bone from arthroscopic imaging, providing both measurement precision and comprehensive bone visualization in one unified view.
Solution Approach 2:
The patent transforms two-dimensional X-ray images into a three-dimensional model, enabling the surgeon to view the entire bone structure from multiple angles rather than being limited to the horizon view provided by traditional X-ray imaging. This allows for comprehensive assessment of bone geometry and pathology.
3Productivity
If traditional two-dimensional imaging is used during surgical procedures, then the imaging can be performed quickly and intraoperatively, but the resulting images provide limited information for assessing joint pathology and surgical progress
Solution Approach 1:
The patent transforms two-dimensional intraoperative images into a three-dimensional model, providing comprehensive joint pathology information and surgical progress assessment while maintaining the speed advantage of intraoperative imaging. The three-dimensional model enables the surgeon to evaluate the entire joint structure, pathology extent, and surgical outcomes in real-time during the procedure.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables surgeons to visualize and measure bone removal more accurately, improving the precision of surgical interventions and enhancing patient outcomes by providing a real-time, three-dimensional snapshot of the joint's state during procedures.
Implementation Method 1
generating three-dimensional image data by back-projecting the first and second imaging data in three-dimensional space
Data Source
AI summary
A method for modeling a joint before, during, and/or after a medical procedure includes receiving first imaging data capturing the joint from a first imaging perspective and second imaging data capturing the joint from a second imaging perspective that is different than the first imaging perspective, the first and second imaging data generated intraoperatively via a two-dimensional imaging modality, generating three-dimensional image data by back-projecting the first and second imaging data in three-dimensional space in accordance with a relative difference between the first and second imaging perspectives, generating a three-dimensional model of the joint based on processing the three-dimensional image data with a machine learning model trained on imaging data generated via at least a three-dimensional imaging modality, and displaying a visualization based on the three-dimensional model of the joint during the medical procedure.