Parametric Image Polarity Trend Analysis for Perfusion Kinetics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Parametric images in diagnostic applications, such as contrast-enhanced ultrasound imaging, fail to effectively represent the dynamic behavior of blood perfusion kinetics, leading to poor results in characterizing lesions like Focal Liver Lesions with Dynamic Vascular Patterns, which are challenging to assess due to their differing perfusion kinetics from healthy tissue.
Innovation Solution
A data-processing method that generates a single parametric image by analyzing a sequence of input images over time, comparing analysis functions at selected locations to a reference function, and calculating a polarity trend to classify and represent perfusion kinetics, allowing for immediate visualization and assessment of perfusion dynamics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If a single parametric value is assigned to each location to simplify visualization, then the ease of operation is improved, but the loss of information about dynamic perfusion behavior increases
Solution Approach 1:
The patent segments the continuous perfusion kinetics data into discrete polarity trend categories (e.g., increasing, decreasing, stable, biphasic). This segmentation allows the complex dynamic behavior to be represented in a simplified categorical manner while preserving the essential kinetic characteristics, resolving the contradiction between simplification and information retention.
Solution Approach 2:
The patent transforms the temporal dynamics of perfusion kinetics into a polarity parameter that captures the direction and pattern of change over time. By changing the representation from continuous temporal data to a polarity parameter with specific categories, the system maintains diagnostic information while enabling straightforward visualization in the parametric image.
2Device complexity
If binary classification is used to determine early wash-in locations, then the device complexity is reduced, but the measurement precision of perfusion kinetics decreases
Solution Approach 1:
The patent introduces dynamic polarity trends that capture the temporal evolution of perfusion kinetics. Instead of static binary classification, the system evaluates how the perfusion signal changes over time (increasing, decreasing, stable, biphasic patterns), providing a more precise characterization of lesion perfusion behavior while maintaining relatively simple classification logic.
3Measurement precision
If animated sequences of computed images are generated to show perfusion evolution, then the measurement precision of perfusion kinetics is improved, but the loss of time increases
Solution Approach 1:
The patent segments the temporal perfusion information into discrete polarity trend categories that can be encoded in a single parametric image. This segmentation allows the essential kinetic information to be captured and visualized immediately without requiring playback of animated sequences, thus reducing analysis time while preserving diagnostic precision.
Solution Approach 2:
The patent encodes temporal dynamics into a new dimension - the polarity trend parameter - that can be represented in the parametric image alongside spatial information. This dimensional transformation allows kinetic information to be visualized statically in the image without requiring temporal animation, eliminating the time loss associated with reviewing image sequences.
4Measurement precision
If detailed curve fitting and comparison are performed for each location, then the measurement precision of perfusion parameters is improved, but the device complexity increases
Solution Approach 1:
The patent extracts only the essential polarity trend information from the complete perfusion curve, rather than processing and storing all curve details. By taking out only the critical kinetic characteristic (the polarity pattern of change), the system maintains measurement precision for lesion characterization while significantly reducing data processing complexity.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A solution for analyzing a body-part of a patient is proposed. A corresponding data-processing method (A1-A14) includes the steps of providing (A1) a sequence of input images representing the body-part over an analysis period, each input image including a set of input values each one being indicative of a response to an interrogation signal of a corresponding location of the body-part at a corresponding acquisition instant included in the analysis period, associating (A2,A3,A4,A6) an analysis function of time with each one of a set of selected locations, the analysis function modeling a trend of the input values of the selected location in the sequence of input images, and providing (A2',A4',A5,A6') a reference function of time for the analysis functions; in the solution according to an embodiment of the invention, the data-processing method further includes comparing (A7) the analysis function of each selected location with the reference function to determine a polarity trend representing a trend over the analysis period of a polarity of a divergence between the analysis function of the selected location and the reference function, and creating (A8-A13) a parametric image including a parametric value for each selected location, the parametric value being indicative of the polarity trend of the selected location.