Finger Pose Registration for Accurate Anatomical Region Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately determine the position and orientation of anatomical regions of interest, such as bones or skulls, for precise registration with medical imaging datasets, which is crucial for navigation in surgical procedures.
Innovation Solution
A method that involves generating a pose model of a user's hand, determining its position relative to an anatomical region of interest, and using this position to register a patient image dataset, allowing for precise alignment with a camera coordinate system.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional surface matching algorithms are used to register anatomical structures, then the registration process can be performed, but the measurement precision and reliability of anatomical region positioning are insufficient for accurate surgical navigation
Solution Approach 1:
The patent introduces a finger model as an intermediary tool between the surgeon and the anatomical structure. The finger model is placed on the anatomical region of interest, and its position is detected by a camera system. This intermediary finger model serves as a bridge to transfer the surgeon's manual positioning into precise digital coordinates, enabling accurate registration without requiring complex automated surface matching algorithms.
2Extent of automation
If automated surface matching algorithms are employed, then the registration process is automated, but the ease of operation and adaptability to different anatomical structures are reduced
Solution Approach 1:
The finger model-based system enables the surgeon to directly control the registration process through their own hand movements. The surgeon places their finger on the anatomical structure and naturally guides the positioning, with the system automatically detecting the finger position and performing the registration. This self-service approach combines manual operator control with automated detection, maintaining ease of operation while achieving registration automation.
3Measurement precision
If invasive surgical activities are performed to establish accurate registration, then the measurement precision improves, but the loss of time and complexity of the procedure increases
Solution Approach 1:
The patent replaces invasive mechanical surgical activities with a non-invasive optical detection system. Instead of performing surgical interventions to establish registration landmarks, the system uses a camera to detect the position of a finger model placed on the anatomical structure. This substitution eliminates the need for invasive procedures while maintaining high measurement precision and reducing procedure time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Disclosed is a computer-implemented medical method of determining the position of an anatomical region of interest of a patient's body, the method comprising the following steps: a) finger model data is acquired (S11) which describes at least one model of a pose of at least one finger of a user; b) finger position data is acquired (S12) based on the finger model data and based on imaging the at least one finger, wherein the finger position data describes a position of the at least one finger at which the at least one finger points at the anatomical region of interest and wherein the finger position data is acquired by extracting, from a camera image generated by the imaging and describing the at least one finger, the position of the image representation of at least part of the at least one finger by comparing the camera image to the model of the at least one finger; c) planning image data is acquired (S13) which describes a planning external surface of the anatomical region of interest; d) anatomical region position data is determined (S14) based on the finger position data and the planning image data, wherein the anatomical region position data describes the position of the anatomical region of interest, wherein the anatomical region position data is determined by determining, based on the finger position data, an actual surface of the anatomical region of interest and comparing the actual external surface to the planning external surface, wherein the actual surface is determined by generating or represented by a point cloud of positions of the at least one finger, wherein the actual external surface is determined to be the planning external surface if the comparison between the actual external surface and the planning external surface results in that the actual external surface is similar to the planning external surface at least to a predetermined degree of similarity, wherein the finger model data describes a user-specific model of the pose which is acquired by imaging the at least one finger when it attains the pose.