Ultrasound Super-Resolution Preprocessing for Microbubble Decoupling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current ultrasound super-resolution imaging techniques face challenges in accurately reconstructing microvasculature images due to high microbubble concentrations and strong noise interference, limiting their clinical application.
Innovation Solution
A super-resolution reconstruction preprocessing method that analyzes grayscale fluctuation signals across multiple frames of contrast-enhanced ultrasound images to distinguish microbubble signals from noise, using interpolation and weighted calculations to enhance image resolution and decouple overlapping microbubbles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If high-concentration microbubbles are used to shorten acquisition time, then acquisition time is reduced, but microbubble overlap increases which affects localization accuracy
Solution Approach 1:
The patent segments the microbubble signals by separating collocated pixel points (same position across frames) from associated pixel points (adjacent positions). This segmentation allows independent processing of microbubble trajectories versus background signals, enabling accurate localization even when microbubbles are concentrated and may overlap in the raw images.
Solution Approach 2:
The patent introduces grayscale fluctuation signals as an intermediary representation. Instead of directly processing the raw ultrasound images, it converts them into temporal fluctuation signals that capture microbubble dynamics. This intermediary transformation enables separation of signal from noise and allows for accurate localization without direct visual overlap interference.
2Device complexity
If standard CEUS super-resolution methods are used, then image processing is simplified, but reconstruction speed and accuracy are insufficient under strong noise interference
Solution Approach 1:
The patent performs preliminary action by pre-processing the ultrasound images to extract and organize grayscale fluctuation signals before the main reconstruction process. It pre-separates collocated and associated pixel points and prepares temporal fluctuation signals, which then enables fast and accurate reconstruction without complex real-time calculations during the actual image generation.
3Reliability
If noise filtering is applied to improve signal quality, then noise is reduced, but processing time increases
Solution Approach 1:
The patent applies local quality by treating different pixel regions differently - collocated pixel points (which contain microbubble signal information) are processed with one set of operations while associated pixel points (containing background information) are processed separately. This localized approach filters noise effectively at each location without requiring global reprocessing, maintaining speed while improving signal quality.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
A super-resolution reconstruction preprocessing method of contrast-enhanced ultrasound images includes: acquiring an image set to be preprocessed; acquiring grayscale fluctuation signal of a pixel point in the registered contrast-enhanced ultrasound images to be preprocessed; performing denoising and reconstruction operation on the image set to be preprocessed to obtain a reconstructed feature parameter image based on grayscale fluctuation signals of a collocated pixel point set, and performing interpolation calculation on the reconstructed feature parameter image to obtain a sparse microbubble image based on the grayscale fluctuation signals of the collocated pixel point set and grayscale fluctuation signals of an associated pixel point set associated with the collocated pixel point set. By analyzing the grayscale fluctuation signals of the collocated pixel point set in the plurality of frames of the registered contrast-enhanced ultrasound images to be preprocessed, a signal-to-noise ratio and a signal-to-background ratio are improved. By performing interpolation operation on the reconstructed feature parameter image using a similarity of the grayscale fluctuation signals of the collocated pixel point set and the grayscale fluctuation signals of the associated pixel point set, spatial decoupling of overlapping microbubbles is realized, and influence of strong noise and high concentration microbubble on reconstruction is reduced.