Presurgical 3D Device Localization From 2D Fluoroscopic Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing 2D medical imaging technologies inadequately represent the position and orientation of invasive medical devices, leading to increased procedure duration and radiation exposure for patients and operators.
Innovation Solution
A system utilizing machine learning algorithms to infer 3D position and orientation of medical devices from 2D images, enhancing existing imaging devices with 3D information without additional hardware, by training models on annotated images to predict device orientation and position.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If 2D imaging technologies are used to guide medical procedures, then practitioners can obtain real-time views of device placement, but the position and orientation information of medical devices is inadequately represented
Solution Approach 1:
The patent applies dimensionality change by inferring three-dimensional position and orientation information from two-dimensional fluoroscopic images. The machine learning model processes 2D image data and outputs 3D spatial coordinates (x, y, z position and roll, pitch, yaw orientation), effectively adding depth and orientation dimensions that are not directly visible in the 2D imaging plane. This resolves the information loss by computationally reconstructing the missing dimensional data.
2Productivity
If practitioners manually determine device position and orientation from 2D images, then no additional hardware is needed, but procedure duration increases
Solution Approach 1:
The patent replaces the manual mechanical process of visual assessment and geometric calculation with an automated machine learning system. Instead of practitioners manually measuring and calculating device position and orientation from 2D images, the system uses trained neural networks to automatically infer 3D spatial information, significantly reducing the time required while maintaining or improving accuracy.
3Measurement precision
If additional tracking hardware is added to provide 3D information, then position and orientation accuracy improves, but device complexity and cost increase
Solution Approach 1:
The patent creates a computational copy or representation of the physical tracking system. Instead of adding physical sensors and markers to the medical device, the system creates a virtual model that infers spatial information from existing 2D image sequences. The machine learning model learns the mapping between 2D image appearances and 3D spatial states, effectively copying the tracking function through software rather than hardware.
4Reliability
If fluoroscopy is used continuously to monitor device placement, then real-time visualization is provided, but radiation exposure to patient and operator increases
Solution Approach 1:
The patent applies partial action by using the machine learning model to infer 3D position and orientation from limited 2D image data rather than requiring continuous or excessive fluoroscopic imaging. The system can accurately track device position using fewer imaging frames, reducing the total radiation dose while maintaining reliable real-time monitoring capability through intelligent image analysis.
Data Source
AI summary
A system, method, and computer program product for presurgical planning related to an object associated with a medical procedure for a patient is disclosed. A model can be trained for an object as part of a medical procedure, the trained model being generated from one or more machine learning algorithms that are trained on annotated images of the object with spatial information of the object. An imaging computer system can receive or simulate one or more images of the object captured by an imaging device positioned external to the patient. The imaging computer system can further determine, based on applying the trained model to the one or more images of the object, spatial information of the object during the medical procedure. A display can output the one or more images and the spatial information of the object.