3D Image Dataset Reconstruction via Multi-Angle Projection Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional medical X-ray devices face limitations in achieving complete 3D imaging due to a limited detector size, which restricts the imaging region and often results in truncation artifacts, particularly in anatomical regions like the lung and liver, and cannot acquire certain anatomical regions with a single recording of projection images.
Innovation Solution
A method involving the recording of multiple first and second projection images from different angles, where each pair of images shares a common stationary center point, allowing for the reconstruction of a 3D image dataset using a medical X-ray device with an X-ray source and detector arranged on a C-arm, enabling the combination of these images to cover larger volumes and reduce radiation burden.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If a single recording of projection images is used, then the imaging process is simple and fast, but the imaging region is limited and cannot cover particular anatomical regions
Solution Approach 1:
The imaging process is segmented into multiple recording passes, each capturing a different angular range. First projection images are recorded with the X-ray source and detector in an initial position, then second projection images are recorded after rotating the C-arm to a different angular position. This segmentation allows the limited detector area to cover a larger overall imaging region by combining multiple recordings.
Solution Approach 2:
The solution extends the imaging capability from a single angular view to multiple angular views by rotating the X-ray source and detector assembly. This adds an angular dimension to the imaging process, allowing projection images to be acquired from different angles and subsequently combined to reconstruct a complete 3D image of larger anatomical regions.
2Area of stationary object
If the detector size is increased to cover larger volumes, then the imaging region expands, but the device size and cost increase
Solution Approach 1:
Instead of using a large stationary detector, the solution employs a smaller detector that is dynamically positioned at different angles around the examination object. The C-arm assembly rotates to acquire projection images from multiple angular positions, allowing a compact detector to achieve the coverage of a much larger detector through dynamic repositioning.
3Area of stationary object
If multiple projection images from different angles are recorded, then complete 3D imaging of larger volumes is achieved, but the radiation exposure increases
Solution Approach 1:
The method uses partial recordings at strategically selected angular positions rather than continuous 360-degree scanning. By acquiring projection images at specific angles that provide sufficient coverage for 3D reconstruction, the radiation exposure is minimized while still achieving complete imaging of the anatomical region of interest.
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
Enables complete 3D imaging of larger volumes compared to conventional methods, reducing truncation artifacts and allowing for more comprehensive anatomical coverage while minimizing radiation exposure.
Implementation Method 1
The X-ray device often has an X-ray source for emitting X-rays for irradiating the examination object, and a detector for receiving the X-rays
Data Source
AI summary
First projection images are recorded. The first projection images map an examination object along different first projection directions. A corresponding respective second projection image is recorded for at least two first projection images. The first projection images each have a first focus point, and the second projection images each have a second focus point. Each of the second projection images together with a corresponding first projection image at least partially map a common part of the examination object about a stationary center point. The second projection images map the examination object along second projection directions that are mutually different and at least partially different, relative to the respectively corresponding first projection directions such that a straight line through the first focus point and the second focus point of the mutually corresponding first projection images and second projection images extends through the stationary center point. The 3D image dataset is reconstructed.