Perfusion Data Reconstruction Using Weighted Base Functions
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for acquiring three-dimensional perfusion data sets are limited by the Nyquist-Shannon sampling theorem, requiring high-frequency recording of projection images, which is not feasible with C-arm X-ray devices, and iterative methods for parameterizing voxel values are computationally intensive.
Innovation Solution
Determining projection weighting factors for each recording geometry and base function as a function of intensity characteristics over time, using interpolation to achieve higher temporal resolution and reducing computational complexity, and calculating volume weighting factors from these projection weighting factors using orthogonality and known recording geometries.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If direct reconstruction methods are used to acquire perfusion data, then three-dimensional perfusion data sets can be obtained, but the time resolution is limited by the recording frequency which must satisfy the Nyquist-Shannon sampling theorem
Solution Approach 1:
The patent applies preliminary action by using prior knowledge about the expected time characteristics of contrast medium concentration (modeled by base functions) before actual reconstruction. This allows the system to prepare appropriate weighting factors and reconstruction parameters in advance, enabling accurate perfusion data acquisition at lower recording frequencies than would otherwise be required by the Nyquist-Shannon theorem.
Solution Approach 2:
The patent changes parameters by introducing projection weighting factors that are determined as a function of intensity characteristics and base functions. These weighting factors are optimized to maximize the information content at lower sampling rates, effectively changing the reconstruction parameters to achieve higher temporal resolution without requiring proportionally higher recording frequencies.
2Measurement precision
If iterative methods are used to determine parameterized voxel values with prior knowledge, then perfusion data can be reconstructed using available recording frequencies, but the computational complexity and computing time increase significantly
Solution Approach 1:
The patent extracts only the essential prior knowledge needed for reconstruction by using a limited set of parameterized base functions that describe the expected time characteristics of contrast medium concentration. Instead of using complex iterative methods that require extensive computation, the invention extracts and utilizes only the necessary temporal patterns, significantly reducing computational complexity while maintaining accuracy.
Solution Approach 2:
The patent transforms the complex iterative parameter determination problem into a simpler parameter optimization problem by pre-defining base functions with specific time characteristics. The reconstruction then becomes a matter of determining weighting factors for these predefined functions rather than performing full iterative optimization, dramatically reducing computational requirements.
3Measurement precision
If high recording frequencies are used to achieve sufficient temporal resolution, then accurate perfusion data can be acquired, but the radiation exposure and acquisition time increase
Solution Approach 1:
By incorporating prior knowledge about contrast medium concentration time characteristics into the reconstruction algorithm through predefined base functions, the system can achieve accurate temporal resolution at lower recording frequencies. This preliminary incorporation of expected patterns allows the system to extract maximum information from fewer measurements, thereby reducing radiation exposure.
Solution Approach 2:
The patent changes the approach from requiring high sampling frequencies to using optimized weighting factors based on intensity characteristics and base functions. This parameter transformation allows accurate perfusion measurement at reduced recording rates, directly reducing the number of X-ray exposures required while maintaining temporal resolution.
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
This approach allows for efficient acquisition and reconstruction of time-dependent three-dimensional perfusion data sets with reduced radiation exposure and computational intensity, enabling higher temporal resolution and robust reconstruction of perfusion data.
Implementation Method 1
an examination object is irradiated with X-ray radiation which has passed at least partially through the examination object
Implementation Method 2
detecting the time characteristic of the contrast medium concentration as a function of location
Data Source
AI summary
A method and system are provided for determining a time-dependent, three-dimensional perfusion data set relating to the perfusion of at least one vessel and/or of tissue of an examination object. Projection images of the vessel and/or tissue are acquired in a plurality of recording geometries by an X-ray detector at a plurality of recording times in each case, which images describe detected intensities in a plurality of imaging regions of the X-ray detector. The perfusion data set is determined by associating a weighted sum of specified time-dependent base functions with each voxel of the perfusion data set.