Contrast Ultrasound Frame Selection Using Parametric Maps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current contrast-enhanced ultrasound imaging techniques require manual frame selection, which is time-consuming and prone to human error, leading to inconsistent and reduced diagnostic value due to incomplete tissue suppression and residual tissue signals.
Innovation Solution
An apparatus and method that analyze grayscale envelope statistical distributions within multi-pixel windows across each contrast image frame to automatically select the first and last frames for generating accumulation images, using features like SNR, NI, and SNR×NI for consistent frame selection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual frame selection is used for generating accumulation images, then the diagnostic value can be maintained through expert judgment, but the process is time-consuming and prone to human error
Solution Approach 1:
The system performs automatic frame selection using computational algorithms that analyze image features independently, without requiring continuous human intervention. The processor automatically identifies optimal frames based on contrast enhancement metrics, allowing the system to serve itself in the frame selection task while maintaining diagnostic quality
Solution Approach 2:
The manual mechanical process of frame selection by human operators is replaced with an automated computational system. The processor uses image feature analysis and statistical methods to substitute human judgment with algorithmic decision-making, eliminating time loss while preserving diagnostic reliability
2Adaptability or versatility
If manual frame selection is used, then flexibility in selection can be maintained, but consistency is reduced due to human error
Solution Approach 1:
The system maintains flexibility by allowing adjustment of image feature parameters and analysis thresholds. Different image features (e.g., contrast enhancement metrics, statistical distributions) can be weighted or selected based on specific diagnostic needs, while the automated processing ensures consistent application of chosen parameters across all frame selections
3Device complexity
If grayscale intensities of individual pixels are used for frame selection, then the process is simple, but tissue signals are not suppressed completely leading to poor differentiation
Solution Approach 1:
The system transitions from analyzing single-pixel grayscale intensities to examining multi-pixel window statistical distributions. This dimensional expansion from point-based to region-based analysis provides richer information about tissue contrast and signal characteristics, enabling better differentiation of contrast echoes while maintaining manageable computational complexity
4Measurement precision
If image features from multi-pixel windows are used for automatic frame selection, then differentiation of contrast echoes improves, but processing complexity increases
Solution Approach 1:
The system extracts specific image features (such as contrast enhancement metrics and statistical distribution parameters) from multi-pixel windows, separating the essential diagnostic information from the full image data. This extraction approach maintains high measurement precision while reducing processing complexity by focusing computation on key extracted features rather than all pixel data
Data Source
Figure 1A~1B
Figure 2
Figure 3
AI summary
Apparatuses, systems, and methods are provided for automatically selecting first and last frames for a sequence of frames from which an accumulation contrast image may be generated. In some examples, statistical distributions of groups of pixels of the image frames may be analyzed to generate parametric maps. The parametric maps may be analyzed to select the first and last image frames of the sequence. In some examples, an image frame corresponding to the parametric map having a value above a threshold value may be selected as a first frame. In some examples, an image frame corresponding to the parametric map having a maximum value of all the parametric maps may be selected as the last frame. In some examples, the parametric maps may be used to segment features, such as a tumor, from the image frame.