Fluoroscopy Surgical Device Tracking via Deformable Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for tracking surgical devices in non-rigid body organs, such as the lung, during minimally invasive procedures face challenges due to differences in patient posture and breathing motion, making real-time tracking and guidance uncertain and unreliable.
Innovation Solution
A method that involves receiving 3D image data of the non-rigid body organ, computing a 3D model, and deforming it to match the organ's real-time position using reference marks, while registering fluoroscopy images to determine the surgical device's 3D location, and providing this information for display to assist in guiding the device to a region of interest.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple 2D X-ray images from multiple fluoroscopic camera views are used to track a surgical device, then the tracking information can be obtained, but the determination of device position relies on physician skill and experience resulting in considerable uncertainty
Solution Approach 1:
The patent replaces the manual mechanical interpretation method (physician visually analyzing 2D X-ray images) with an automated computer-based image processing system. The system automatically detects reference marks in fluoroscopic images, computes their 3D positions, and tracks the surgical device without relying on physician skill or experience, thereby eliminating the considerable uncertainty in position determination.
Solution Approach 2:
The patent creates a virtual 3D model that copies and represents the physical surgical scene. By detecting reference marks and computing their 3D positions, the system creates a digital twin of the surgical environment, allowing accurate device position determination through computational geometry rather than manual interpretation of 2D images.
2Measurement precision
If a 3D reconstructed image is registered with 2D X-ray images using rigid registration technique, then visualization of rigid structures like heart and spine is improved, but non-rigid body organs such as lung cannot be accurately visualized due to patient position changes and breathing motion
Solution Approach 1:
The patent transitions from rigid registration to dynamic deformable registration. The system models the lung as a deformable organ that changes shape and position during breathing and patient movement. By using deformable registration techniques, the system can dynamically adapt the 3D model to match the actual organ position and shape in real-time fluoroscopic images, enabling accurate tracking of surgical devices in non-rigid organs.
Solution Approach 2:
The patent changes the registration parameters from rigid transformation (translation and rotation only) to deformable transformation that includes local shape changes. This allows the 3D model to accommodate breathing motion and position changes by adjusting tissue deformation parameters, making the system applicable to non-rigid organs like the lung while maintaining visualization accuracy.
3Loss of time
If pre-operative 3D image data is acquired when the patient is in one position, then the 3D model can be created, but the model does not match the real-time organ position when the patient is in a different position during surgery
Solution Approach 1:
The patent performs preliminary acquisition of 3D image data (such as CT or MRI scans) before the surgical procedure when the patient is in a planning position. This allows adequate time for high-quality image acquisition and 3D model construction without rushing the process. The model is then adapted to the surgical position using deformable registration, resolving the conflict between taking time for accurate pre-operative imaging and maintaining accuracy during position changes.
Solution Approach 2:
The patent makes the 3D model dynamic by applying deformable registration that adapts the pre-operative model to the patient's actual position during surgery. This dynamic adaptation compensates for position changes and breathing motion, maintaining model-organ matching accuracy throughout the procedure without requiring repeated time-consuming image acquisitions.
Data Source
AI summary
A method for assisting a physician track a surgical device in a body organ of a subject during a procedure includes fluoroscopic based registration, and tracking. An initial registration step includes receiving a 3D image data of a subject in a first body position, receiving a real time fluoroscopy image data, and estimating a deformation model or field to match points in the real time fluoro image with a corresponding point in the 3D model. A tracking step includes computing the 3D location of the surgical device based on a reference mark present on the surgical device, and displaying the surgical device and the 3D model of the body organ in a fused arrangement.