Depth-Encoded Fiducial Marker for Surgical Image Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for image-guided spinal navigation face challenges in registering bony surface anatomy with optical tracking systems, particularly due to user variability, deformation of physical styluses, and the time-consuming nature of exposing clean bone, leading to difficulties in maintaining registration during surgery.
Innovation Solution
The use of a depth-encoded marker that is imageable by multiple imaging systems, allowing for the transformation mapping of intraoperative image data, preoperative image data, and tracking data into a common coordinate space, facilitating accurate and efficient registration through its asymmetry in the depth dimension.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If touch points are collected using a physical stylus, then registration can be accomplished, but user variability and stylus deformation cause loss of registration accuracy
Solution Approach 1:
The patent replaces the mechanical stylus-based touch point collection system with an optical/image-based detection system. The depth-encoded marker is imaged by imaging systems (e.g., intraoperative imaging systems) that automatically determine 3D position and orientation without physical contact, eliminating stylus deformation and user variability while maintaining registration accuracy.
Solution Approach 2:
The patent uses image data copies of the depth-encoded marker from multiple imaging systems to establish registration. Instead of direct mechanical measurement, the system creates and compares digital representations (image data) of the marker's position and orientation across different coordinate spaces, enabling accurate registration without physical interference.
2Productivity
If anatomical points are detected and matched to image set surface contour, then registration can be achieved, but clean exposed bone is required which is time-consuming to achieve
Solution Approach 1:
The patent introduces a depth-encoded marker as an intermediary object that can be imaged by multiple imaging systems. This marker serves as a common reference that bridges intraoperative and preoperative image data, eliminating the need to directly match complex anatomical surface contours and reducing preparation requirements to simple marker placement.
Solution Approach 2:
The patent changes the registration approach from matching complex anatomical surface contours to detecting the 3D position and orientation parameters of a standardized depth-encoded marker. This parameter transformation simplifies the registration process by converting a complex surface-matching problem into a more manageable marker-detection problem with defined geometric parameters.
3Reliability
If conventional registration methods are used, then initial registration can be completed, but re-registration is difficult or impossible if registration is lost during surgery
Solution Approach 1:
The patent places the depth-encoded marker on the patient's anatomy before surgery begins and ensures it remains in place throughout the procedure. This preliminary placement of a stable, imaged marker enables quick re-registration if registration is lost, as the marker provides a permanent reference that can be re-detected and re-used without requiring re-preparation of anatomical landmarks.
4Measurement precision
If multiple markers are used for registration, then registration accuracy can be improved, but device complexity and time to complete registration increase
Solution Approach 1:
The patent designs the depth-encoded marker to serve multiple functions simultaneously: it provides 3D position information, 3D orientation information, and serves as a common reference for multiple different imaging systems (intraoperative and preoperative). This multi-functionality eliminates the need for separate markers for different purposes, reducing overall system complexity while maintaining high registration accuracy.
Solution Approach 2:
The patent merges the functions of multiple potential markers into a single depth-encoded marker that can be detected by multiple imaging modalities. By combining position encoding, orientation encoding, and multi-system detectability into one integrated marker design, the system reduces the number of components needed while achieving comprehensive registration capability.
Data Source
AI summary
Methods and systems of performing intraoperative image registration during a medical procedure are described. A depth-encoded marker is provided to an object of interest. The marker is imageable by at least two imaging systems. The marker has asymmetry in at least a depth dimension. First and second sets of image data are obtained. Tracking data is obtained. The first image coordinate space, the second image coordinate space and the tracking coordinate space are independent from each other. Transformation mapping is performed to register the first and second sets of image data and the tracking data to each other. The first and second sets of image data are mapped to each other based on the depth-encoded marker. The first or second set of image data and the tracking data are mapped to each other based the same or different marker.