Cardiac Strain Trace Correction for Pericardium-Free ROI Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cardiac ultrasound imaging systems inaccurately calculate strain values due to inclusion of non-contractile pericardium and image artifacts, leading to underestimation of healthy regions and increased diagnostic time and effort.
Innovation Solution
The system detects non-physiological strain traces by partitioning the region of interest into segments, analyzing motion vectors, and correcting ROI positioning to exclude non-contractile pericardium and artifacts, using speckle tracking algorithms to improve strain value accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Area of stationary object
If the region of interest (ROI) includes pericardium and image artifacts, then the coverage area is increased, but the strain value accuracy deteriorates due to inclusion of non-contractile tissues
Solution Approach 1:
The patent divides the ROI into multiple segments and further segments each segment into multiple sub-segments based on motion vector analysis. This segmentation allows the system to identify and exclude non-contractile portions (pericardium and artifacts) while retaining contractile myocardium regions, thus maintaining measurement precision while preserving adequate coverage area for reliable strain calculation.
Solution Approach 2:
The patent extracts and removes non-contractile portions from the ROI by analyzing motion vectors and identifying segments with abnormal motion patterns characteristic of pericardium and artifacts. This extraction process eliminates harmful elements while preserving the contractile myocardium regions needed for accurate strain measurement.
2Measurement precision
If manual ROI adjustment is performed to exclude non-contractile pericardium, then the strain value accuracy is improved, but the diagnostic time and user effort increase
Solution Approach 1:
The patent implements automated detection and correction of non-contractile portions using motion vector analysis and segment-based processing. The system performs self-correction by automatically identifying segments containing pericardium or artifacts and adjusting the ROI accordingly, eliminating the need for manual user intervention while maintaining high strain value accuracy.
Solution Approach 2:
The patent uses motion vector feedback from speckle tracking to automatically detect and correct ROI positioning errors. The system continuously monitors motion patterns, compares them against expected physiological ranges, and automatically adjusts the ROI when non-contractile portions are detected, providing real-time feedback-based correction without requiring additional user time.
3Reliability
If multiple repeated imaging acquisitions are performed to obtain accurate strain values, then the measurement reliability is improved, but the productivity and diagnostic efficiency decrease
Solution Approach 1:
The patent performs preliminary detection and correction of non-contractile portions within the initial imaging acquisition by analyzing motion vectors and segmenting the ROI. This preliminary action ensures that accurate strain values are obtained from the first acquisition without requiring repeated imaging, thus maintaining measurement reliability while preserving diagnostic efficiency.
Solution Approach 2:
The patent replaces the mechanical process of repeated physical re-acquisition with an automated computational system that detects and corrects ROI positioning errors through motion vector analysis. This substitution eliminates the need for repeated imaging while maintaining measurement reliability, significantly improving diagnostic efficiency and productivity.
Data Source
AI summary
Systems are herein provided for detection and correction of non-physiological strain traces. In one example, a method comprises generating cardiac ultrasound images from ultrasound imaging data of a heart, generating a segmented region of interest (ROI) of the cardiac ultrasound images, identifying a plurality of points within the segmented ROI, identifying one or more of the plurality of points that correspond to one or more sources of non-physiological strain, correcting the one or more of the plurality of points to generate a corrected segmented ROI with a corrected plurality of points, calculating strain values for the corrected segmented ROI.


