Hollow Organ X-Ray Overlay Registration Using Tomosynthesis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for visual support during medical interventions on hollow organs, such as abdominal aortic aneurysm treatment, face challenges in accurately identifying the aorta without contrast medium and require significant effort in generating 3D image datasets.
Innovation Solution
A method using tomosynthesis to register a pre-operative reference representation with a hollow organ, generating an overlay image by determining registration data without the need for a complete CT reconstruction or contrast medium, and using planning information to define the reference volume.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a complete CT reconstruction is performed to generate 3D image dataset for visual support, then the aorta and planning information can be accurately identified and superimposed, but the effort and time required for generating the 3D image dataset increases significantly
Solution Approach 1:
The patent extracts only the necessary 2D projection images at specific angles (0°, 45°, 90°, 135°) from the complete CT dataset, rather than using all 3D data. This selective extraction reduces processing requirements while maintaining sufficient accuracy for visual support during the intervention.
Solution Approach 2:
The patent uses a partial set of projection images (4 specific angles) instead of the full 3D dataset. This partial action provides sufficient registration accuracy for the intervention while significantly reducing the computational effort and time required compared to complete CT reconstruction.
2Difficulty of detecting and measuring
If contrast medium is used to improve aorta visibility in X-ray images, then the aorta can be clearly identified, but the complexity of the procedure and potential harm to the patient increases
Solution Approach 1:
The patent uses pre-operative CT images as an intermediary reference. These CT images provide clear anatomical information about the aorta and are registered with the fluoroscopy images, serving as a mediator that enables aorta identification without requiring contrast medium during the actual intervention.
Solution Approach 2:
The patent performs the detailed imaging and planning work in advance (pre-operatively) when contrast medium can be used safely. The pre-operative CT provides all necessary anatomical information, which is then used during the intervention without requiring additional contrast medium administration.
3Measurement precision
If many 2D images are recorded from different directions to reconstruct 3D dataset, then accurate registration can be achieved, but the device complexity and operational effort increases
Solution Approach 1:
The patent segments the continuous 3D space into four specific angular positions (0°, 45°, 90°, 135°) and records images only at these discrete angles. This segmentation approach provides sufficient registration accuracy while dramatically reducing the complexity compared to continuous 3D scanning.
Solution Approach 2:
The patent applies different imaging strategies to different parts of the procedure: pre-operative CT provides comprehensive 3D information, while intra-operative imaging uses only four specific 2D projections. This local quality approach optimizes the balance between accuracy and complexity for each stage.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Accurately registers overlay information on X-ray projection images with reduced effort, providing precise visual support during medical interventions without the need for contrast medium or full CT reconstruction.
Implementation Method 1
An X-ray projection image of the hollow organ is generated (e.g., during the intervention, such as by actuation of an X-ray imaging modality by the at least one computing unit)
Implementation Method 2
A reconstruction of a volume of the patient corresponding to the reference volume is generated (e.g., by actuation of the X-ray imaging modality by the at least one computing unit) using a tomosynthesis method
Data Source
AI summary
For visual support in a medical intervention on a hollow organ of a patient, reference data that includes a reference representation of the hollow organ is obtained, an X-ray projection image of the hollow organ is generated, and items of overlay information are determined based on the reference data. The reference data includes items of planning information relating to a location of a characteristic feature of the hollow organ in the reference representation. A reference volume is determined in the reference representation as a function of the items of planning information. A reconstruction of a volume of the patient corresponding to the reference volume is generated using a tomosynthesis method. Registration data is generated in that the reference volume is registered in relation to the reconstruction and the items of overlay information are overlaid on the X-ray projection image as a function of the registration data.