3D CEUS Analysis Using Image Registration for Tissue Motion Compensation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Quantitative analysis of contrast-enhanced ultrasound (CEUS) imaging is challenging due to tissue motion, requiring manual identification of the region of interest (ROI) or volume of interest (VOI) in each ultrasound frame, making the process time-consuming and user-dependent, limiting its application for real-time guidance in biopsy or therapeutic procedures.
Innovation Solution
The method employs image-based registration to automatically compensate for tissue motion in real-time, integrating an ultrasound probe with an external position tracking device to stabilize the image registration and enable automatic identification of ROIs/VOIs, allowing simultaneous acquisition and registration of CEUS and tissue images, which are then used for navigation and guidance in interventional procedures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification of ROI/VOI is performed in each ultrasound frame, then quantitative analysis can be carried out, but the process becomes time-consuming and user-dependent
Solution Approach 1:
The system performs preliminary action by automatically identifying and tracking the ROI/VOI across all ultrasound frames using image-based registration and position tracking, eliminating the need for manual identification in each frame. The ROI is defined once in the baseline image and automatically transformed to subsequent frames through registration algorithms.
Solution Approach 2:
The system implements self-service by using the tissue image itself to guide the registration and ROI transformation process. The tissue image serves as a reference that automatically aligns with contrast images through image-based registration, allowing the system to self-correct for tissue motion without manual intervention.
2Measurement precision
If manual ROI identification is required in each frame, then quantitative analysis is possible, but the method cannot be used for real-time guidance of biopsy or therapeutic procedures
Solution Approach 1:
The system performs preliminary registration between the tissue image and baseline contrast image, establishing a transformation matrix that is then applied in real-time to track the ROI across subsequent contrast frames. This preliminary setup enables rapid real-time analysis without repeated manual intervention.
Solution Approach 2:
The patent replaces the mechanical/manual process of ROI identification with an automated computational system using image-based registration and position tracking. The transformation matrices and registration algorithms automatically compute ROI positions, enabling real-time guidance capability.
3Extent of automation
If image-based registration is used to compensate for tissue motion, then automatic ROI identification is enabled, but the system complexity increases
Solution Approach 1:
The patent introduces an intermediary tissue image that serves as a stable reference between the contrast images and the ROI definition. This intermediary facilitates automatic registration by providing a consistent anatomical reference that is less affected by contrast agent dynamics, simplifying the overall registration process.
Solution Approach 2:
The tissue image serves multiple functions: it acts as a reference for image-based registration, provides anatomical context for ROI definition, and enables position tracking across frames. This multi-functionality reduces the need for separate systems for each task, managing complexity through consolidation.
4Extent of automation
If simultaneous acquisition of CEUS and tissue images is performed, then automatic registration is enabled, but the imaging protocol becomes more complex
Solution Approach 1:
The patent merges the acquisition of CEUS and tissue images into a unified imaging protocol where both image types are acquired simultaneously or in close succession. This combining of acquisition processes enables automatic registration by ensuring temporal and spatial correspondence between the image sets, reducing the need for separate alignment procedures.
Data Source
AI summary
A method (50) for quantitative 3D contrast enhanced ultrasound (CEUS) analysis includes acquiring (54) an initial pair of ultrasound contrast and tissue images of an anatomy. A region of interest (ROI) or volume of interest (VOI) is established (56) in the initial acquired tissue image, which becomes the baseline tissue image. The established ROI/VOI is automatically registered (58) from the initial tissue image to the initial contrast image, which becomes a baseline contrast image. Quantitative analysis is performed (60) on the ROI/VOI of the baseline contrast image. The method further includes acquiring (62) a next ultrasound contrast and tissue image pair, corresponding to an i th current contrast and tissue image pair. Frame-to-frame registration is established (64) between (i) the current tissue image and (ii) the baseline tissue image, the frame-to-frame registration being used (66) between the current tissue image and the baseline tissue image to transfer the ROI/VOI from (i) the baseline contrast image to (ii) the current contrast image, thereby creating a transformed ROI/VOI in the current contrast image. Quantitative analysis is then performed (68) on the transformed ROI/VOI of the current contrast image.