Surgical Navigation Pose Comparison for Obscured Marker Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current surgical navigation systems using RMS error for image registration of fiducial markers are not clinically relevant, as they fail to detect significant offsets in surgical instrument tips due to compensation by non-obscured markers, leading to potential inaccuracies in tracking partially obscured markers.
Innovation Solution
A surgical navigation system that uses a data processor to generate sets of patterns from fiducial markers, compute poses in a reference coordinate system, and apply criteria to detect partially obscured markers by comparing poses between patterns, enhancing the reliability of detecting obscured markers with clinical significance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If RMS error criterion is used for image registration, then the registration can be completed when error is below threshold, but the system fails to detect significant offsets in surgical instrument tips due to compensation by non-obscured markers
Solution Approach 1:
The patent segments the fiducial markers into two distinct groups: fully visible markers and partially obscured markers. By treating these groups separately in the pose computation process, the system can identify which markers are compromised and exclude them from the calculation, thereby preventing the compensation effect that masks tip offsets. This segmentation enables the system to maintain registration completion while improving detection accuracy.
Solution Approach 2:
The patent introduces an intermediary criterion that compares poses computed from different subsets of fiducial markers. This intermediary comparison mechanism acts as a mediator to detect inconsistencies that indicate partially obscured markers, thereby revealing tip offsets that would otherwise be masked by the RMS error compensation effect.
2Device complexity
If RMS error criterion is used for image registration, then the registration process is simplified, but the clinical relevance is compromised due to undetected marker obscuration
Solution Approach 1:
The patent applies preliminary action by first identifying and classifying fiducial markers as fully visible or partially obscured before completing the image registration. This preliminary classification step ensures that compromised markers are excluded from the pose computation, thereby maintaining clinical relevance without significantly increasing overall system complexity.
Solution Approach 2:
The patent implements a feedback mechanism where the system continuously monitors the visibility status of fiducial markers and adjusts the pose computation accordingly. When partially obscured markers are detected, the system provides feedback to exclude these markers from the calculation, thereby maintaining reliable and clinically relevant navigation information throughout the surgical procedure.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A surgical navigation system, comprising: - a surgical tool (1), having a reference zone (Zref); - an optical tracker (T), arranged to track a pose of the surgical tool (1), and having fiducial markers (M1-M8); - a stereoscopic camera (2), arranged to capture images (20) of Ntot fiducial markers (M1-M8), where Ntot is an integer superior or equal to 4; - a data processor (DP), configured to perform steps: a) identify the Ntot fiducial markers (M1-M8) and determine the positions of the Ntot fiducial markers (M1-M8), from the images (20) captured by the stereoscopic camera (2); b) generate a set of patterns (P1-P3), each pattern (P1-P3) having M fiducial markers, where M is an integer superior or equal to 3 and strictly inferior to Ntot; the set of patterns (P1-P3) including N fiducial markers (M1-M8), where N is an integer inferior or equal to Ntot; c) for each pattern (P1-P3) of the set generated in step b): - extract the positions of the corresponding M fiducial markers (M1-M8) thereof from step a); - compute a pose of the reference zone (Zref), in a reference coordinate system, from the extracted positions of the corresponding M fiducial markers (M1-M8); d) apply a criterion to compare the poses of the reference zone (Zref) computed in step c) therebetween for the set of patterns (P1-P3), the criterion being designed to detect a presence of at least one partially-obscured fiducial marker (Mpo) within the N fiducial markers (M1-M8).