CT Perfusion Parameter Maps via Time Normalization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current CT perfusion imaging methods face challenges in accurately obtaining perfusion parameter maps due to differences in image acquisition times across frames, leading to low accuracy and difficulty in solving for the residual function k(t), which affects the temporal resolution and reliability of cerebral blood flow and volume measurements.
Innovation Solution
The method involves pre-processing CT perfusion images to obtain discrete contrast agent concentration and arterial input functions, reading and adjusting acquisition time information to create relative time arrays, fitting or interpolating these curves to align them at a consistent time interval, and re-discretizing to generate accurate CT perfusion imaging parameter maps.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If images of different frames with different acquisition times are used directly to calculate the residual function, then the calculation process is simple, but the accuracy of perfusion parameter maps decreases and the temporal resolution of the residual function becomes low
Solution Approach 1:
The patent applies preliminary action by performing time normalization on the arterial input function and tissue time-density curves before deconvolution calculation. The method pre-processes the time-concentration curves by interpolating them to a common time grid, ensuring that all temporal information is aligned before the residual function calculation. This preliminary time alignment resolves the contradiction by preparing the data in advance, allowing accurate perfusion parameter maps to be generated without requiring complex real-time adjustments during the deconvolution process.
2Reliability
If images of different frames with different acquisition times are used directly, then the processing steps are fewer, but the temporal resolution of the residual function k(t) becomes low
Solution Approach 1:
The patent applies parameter changes by transforming the time parameter of the arterial input function and tissue time-density curves. The method changes the time sampling parameters through interpolation to a standardized time grid, which improves the temporal resolution of the residual function. By modifying the time parameter representation rather than the underlying physiological data, the patent achieves high temporal resolution while maintaining processing efficiency through systematic parameter transformation.
3Measurement precision
If direct deconvolution is performed on original images with different acquisition times, then the calculation is faster, but the accuracy and reliability of perfusion measurements decrease
Solution Approach 1:
The patent introduces an intermediary step of time normalization through interpolation between the original images with different acquisition times and the deconvolution process. The interpolated time-concentration curves serve as a mediator that bridges the temporal discrepancies between frames while preserving the essential physiological information. This intermediary transformation ensures accurate cerebral blood flow and volume measurements without requiring complex direct processing of misaligned original images.
Data Source
AI summary
The application discloses a method, a device, a system and a computer storage medium for obtaining the CT perfusion imaging parameter maps of brain. The method includes: obtaining CT perfusion images, pre-processing the CT perfusion images, and obtaining discrete contrast agent concentration curve C(n) of each pixel point in the brain tissue; reading the acquisition time information of the CT perfusion images to obtain the acquisition time array T(n); intercepting the acquisition time array T(n) to obtain the relative acquisition time array t(n); combining the discrete contrast agent concentration curve C(n) with the corresponding relative acquisition time array t(n) to obtain the discrete time-concentration curve C(tn) of each pixel point in the brain tissue; after fitting or interpolating the discrete time-concentration curve C(tn), re-discretizing at the same time interval, and obtaining the discrete time-concentration curve C(n)′ of each pixel point in brain tissue. The same processing is performed on the arterial input change curve AIF(n) and the venous output change curve VOF(n). The application improves the usability of the tissue density time curve, which helps to reduce the difficulty of solving and improve the accuracy of solving.