Structure Masking for Accurate Device-to-Image Registration
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current minimally invasive medical procedures, such as ablation and biopsy, face challenges in accurate and efficient device-to-image registration due to manual planning, organ movement, deformation, and noise in medical images, leading to increased procedure time and potential tissue damage.
Innovation Solution
A medical guidance device and method that employs structure masking and pre-procedure planning to improve device-to-image registration by suppressing noise and enhancing desired image regions, allowing precise placement of needles or probes in desired geometric configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual registration methods are used for needle placement, then clinicians can perform procedures with basic visualization software, but the process is error-prone and time-consuming
Solution Approach 1:
The patent replaces manual mechanical registration methods with an automated computer-based system that performs device-to-image registration. The system automatically detects fiducial markers in medical images and calculates transformation matrices to align device coordinates with image coordinates, eliminating manual visual estimation and reducing both errors and time consumption.
Solution Approach 2:
The registration system performs self-alignment by automatically detecting fiducial markers and computing transformation parameters without requiring manual intervention. The system independently completes the registration process by matching detected marker positions with their known coordinates, thereby improving efficiency and reliability.
2Measurement precision
If manual visual estimation is used to determine coverage zones, then clinicians can plan ablation procedures, but the estimates are inaccurate and subjective
Solution Approach 1:
The patent replaces subjective visual estimation with an automated computer-based measurement system. The system objectively calculates coverage zones by processing medical images and device coordinates through algorithms that determine precise spatial relationships, eliminating human subjectivity and improving measurement accuracy.
Solution Approach 2:
The system creates accurate digital representations (copies) of the physical anatomy and device positions from medical images. By working with these precise digital models rather than visual estimates, the system achieves superior measurement precision for coverage zone determination.
3Productivity
If fiducial markers are used for automatic registration, then registration speed improves, but noise and artifacts in medical images can cause misdetection
Solution Approach 1:
The patent implements a feedback mechanism where the system detects potential fiducial markers, validates them against expected characteristics, and adjusts detection parameters based on results. This iterative process allows the system to distinguish true fiducials from noise and artifacts, maintaining high detection accuracy while preserving fast automatic registration.
Solution Approach 2:
The system introduces intermediary validation steps between image acquisition and final registration. These intermediate checks analyze detected features for consistency with fiducial marker characteristics, filtering out false positives caused by noise or artifacts before they affect registration accuracy.
Data Source
AI summary
One or more devices, systems, methods and storage mediums for performing medical procedure (e.g., needle guidance, ablation, biopsy, etc.) planning and/or performance, and/or for performing registration using at least one mask, are provided. Examples of applications for such devices, systems, methods and storage mediums include imaging, evaluating and diagnosing biological objects, such as, but not limited to, lesions and tumors, and such devices, systems, methods and storage mediums may be used for radiotherapy applications (e.g., to determine whether to place seed(s) for radiotherapy). The devices, systems, methods and storage mediums provide improved registration results by utilizing at least one mask to suppress one or more artifacts or objects (which may or may not include, but is not limited to, at least one medical instrument or tool) in an image including a portion of the medical guidance device and/or to enhance a region or target of interest in the image.


