Virtual Contrast Image Necrotic Tissue Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining the proportion of necrotic tissue in tumors are inefficient due to the need for exact alignment and adaptation of image datasets, which is challenging due to patient movements and deformations, and require significant computer resources, making them impractical for clinical use.
Innovation Solution
A method using high-energy and low-energy image datasets recorded with different x-ray energies after contrast medium injection, which allows for the creation of a virtual contrast medium image, segmentation of tissue areas, and analysis to identify necrotic tissue without the need for a native image reference, reducing computational effort and susceptibility to movement artifacts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If subtraction method is used to determine necrotic tissue from two image datasets, then necrotic tissue proportion can be determined, but exact alignment and adaptation of image datasets is required which is challenging due to patient movements and deformations
Solution Approach 1:
The patent applies preliminary action by performing the subtraction of image datasets immediately after contrast medium injection, before significant patient movement or tissue deformation can occur. This timing strategy ensures that the anatomical structures remain in nearly identical positions across the two energy acquisitions, eliminating the need for complex non-rigid registration algorithms while maintaining accurate necrotic tissue differentiation.
2Measurement precision
If non-rigid registration is used to compensate for tissue displacements, then alignment accuracy is improved, but large computer capacities are required and results are not straightforward
Solution Approach 1:
The invention performs the critical subtraction operation immediately after contrast injection, before physiological movements and tissue deformations have time to develop. This preliminary timing action renders complex non-rigid registration unnecessary, as the anatomical structures remain sufficiently stable, thereby achieving both alignment accuracy and computational efficiency.
3Loss of information
If two recordings with time interval are made to allow contrast medium accumulation, then necrotic tissue differentiation is possible, but patient movements and tissue deformations occur between recordings
Solution Approach 1:
The patent performs both low-energy and high-energy recordings in immediate succession after contrast medium injection, capturing the contrast accumulation state at essentially the same moment. This eliminates the temporal gap that would otherwise allow patient movement and tissue deformation, while still allowing sufficient contrast accumulation to occur during the injection and immediate post-injection period.
4Measurement precision
If native recording without contrast medium is used as reference, then necrotic tissue can be identified, but the method requires significant computational resources and is hardly used in clinical practice
Solution Approach 1:
The invention performs both low-energy and high-energy contrast-enhanced recordings immediately after contrast injection, before anatomical changes occur. This allows direct subtraction between the two energy datasets to generate a virtual non-contrast image, eliminating the need for a separate native scanning session and the associated complex registration requirements, thereby improving clinical workflow efficiency while maintaining diagnostic accuracy.
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 provides improved accuracy and reduced computational requirements, enabling a more reliable and efficient determination of necrotic tissue proportion in tumors, enhancing the assessment of therapy success and reducing the risk of removing necrotic tissue during biopsies.
Implementation Method 1
high-energy image dataset comprising the tissue area of the object under examination and a low-energy image dataset comprising the tissue area of the object under examination which have been created by means of x-ray measurements with different x-ray energies
Data Source
AI summary
An image processing device and method are disclosed for determining a proportion of necrotic tissue in a defined tissue area of an object under examination based on a high-energy image dataset and a low-energy image dataset, each recorded by way of x-ray measurements with different x-ray energies after a contrast medium has been applied to the object under examination. In at least one embodiment of the method, a virtual contrast medium image is determined from the high-energy image dataset and the low-energy image dataset and a segmentation image dataset is created, by the area of tissue being segmented. The segmentation result is transferred into the virtual contrast medium image for segmenting the tissue area in the virtual contrast medium image. Finally an analysis of values of the pixels lying in the segmented area is undertaken for identifying pixels which are to be assigned to necrotic tissue.