Motion-guided Segmentation for Cine DENSE Myocardial Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current MRI techniques for myocardial tissue tracking, such as cine DENSE, are limited by the labor-intensive process of manually delineating myocardial contours at all cardiac phases, and lack automation due to challenges like indiscernible tissue boundaries, signal-to-noise ratio decay, and high signal in blood, which hinders efficient diagnosis and management of heart disease.
Innovation Solution
A method using spatiotemporal phase unwrapping and displacement vector analysis to project a single manually defined set of myocardial contours through time, employing a modulus deformation mask and 2D interpolation with Gaussian functions to smooth and threshold images, thereby isolating and tracking myocardial motion trajectories.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual delineation of myocardial contours is performed at all cardiac phases, then accurate segmentation is achieved, but the process becomes labor-intensive and time-consuming
Solution Approach 1:
The patent performs manual contour delineation only once at a reference cardiac phase (e.g., end-diastole) instead of at all phases. This preliminary manual action serves as the foundation for automatically generating contours at all other cardiac phases through motion tracking and phase unwrapping techniques, dramatically reducing the time required while maintaining segmentation accuracy.
Solution Approach 2:
The patent creates copies of the manually delineated reference contour and transforms them to match the myocardial geometry at different cardiac phases. By using phase unwrapping and motion tracking, the reference contour is effectively copied and adapted to represent the myocardium at systole and intermediate phases, eliminating the need for repeated manual delineation.
2Extent of automation
If automated contour detection based on image intensity is used, then the process is automated, but it fails due to indiscernible boundaries, signal decay, and high blood signal
Solution Approach 1:
The patent introduces phase unwrapping and motion tracking as intermediary processes between the reference contour and the contours at other phases. These intermediaries use the encoded phase information from cine DENSE images, which contains explicit motion data, to accurately transform the reference contour to match myocardial geometry at different phases, overcoming the limitations of direct intensity-based detection.
Solution Approach 2:
The patent changes from using image intensity parameters (which are unreliable due to decay and blood signal) to using phase parameters from cine DENSE images. The phase information encodes myocardial motion directly and provides discernible boundaries throughout the cardiac cycle, enabling accurate automated contour detection at all phases without the limitations of intensity-based methods.
3Ease of operation
If a single reference contour is manually defined, then user interaction is minimized, but accurate projection through time requires sophisticated motion tracking
Solution Approach 1:
The patent replaces complex manual contouring operations (mechanical user interaction) with automated computational processes. The motion tracking system uses phase unwrapping algorithms and displacement vector calculations to automatically transform the reference contour through time, substituting the need for repeated manual operations with a sophisticated but automated computational pipeline.
Data Source
AI summary
Myocardial tissue tracking techniques are used to project or guide a single manually-defined set of myocardial contours through time. Displacement encoding with stimulated echoes (DENSE), harmonic phase (HARP) and speckle tracking is used to encode tissue displacement into the phase of complex MRI images, providing a time series of these images, and facilitating the non-invasive study of myocardial kinematics. Epicardial and endocardial contours need to be defined at each frame on cine DENSE images for the quantification of regional displacement and strain as a function of time. The disclosed method presents a novel and effective two dimensional semi-automated segmentation technique that uses the encoded motion to project a manually defined region of interest through time. Contours can then easily be extracted for each cardiac phase.