Related Experiment Video
Updated: May 26, 2026
11:16
In vitro Assessment of Aortic Regurgitation Using Four-Dimensional Flow Magnetic Resonance Imaging
Published on: February 25, 2022
Mitral annulus segmentation from four-dimensional ultrasound using a valve state predictor and constrained optical
Robert J Schneider1, Douglas P Perrin, Nikolay V Vasilyev
1Harvard School of Engineering and Applied Sciences, Cambridge, MA, USA. rjschn@seas.harvard.edu
Medical Image Analysis
|December 28, 2011
Summary
This study introduces a novel four-dimensional (4D) ultrasound method for precise mitral valve annulus segmentation. The technique accurately tracks annulus motion and shape, improving diagnostic and modeling applications.
Area of Science:
- Medical imaging
- Biomedical engineering
- Cardiovascular research
Background:
- Accurate measurement of the mitral valve annulus shape and motion is crucial for diagnosing cardiac pathologies and developing mitral valve models.
- Existing four-dimensional (4D) ultrasound methods for annulus segmentation suffer from limitations such as high user interaction, cumulative tracking errors, and failure to account for dynamic annular shape and motion.
Purpose of the Study:
- To present a novel 4D annulus segmentation method that overcomes the deficiencies of current techniques.
- To provide an accurate and robust method for delineating the mitral valve annulus throughout the cardiac cycle using 4D ultrasound.
Main Methods:
- The method builds upon a validated three-dimensional (3D) segmentation algorithm for closed-valve frames.
- A valve state predictor identifies frames with a closed mitral valve for 3D segmentation.
- A constrained optical flow algorithm tracks the annulus in open-valve frames, requiring minimal user input (one closed-valve frame and one reference point).
Main Results:
- The new 4D method accurately segments the mitral annulus by combining 3D segmentation during valve closure and optical flow tracking during opening.
- The algorithm demonstrates robustness and requires minimal, imprecise user initialization.
- Comparison with expert manual segmentations across 30 frames yielded an average Root Mean Square (RMS) difference of 1.67±0.63mm, validating the tracking accuracy.
Conclusions:
- The proposed 4D annulus segmentation method offers a significant advancement over existing techniques by accurately capturing dynamic annular shape and motion.
- This method provides a robust, user-friendly, and accurate tool for clinical applications like cardiac pathology diagnosis and computational modeling.
- The minimal user interaction and high accuracy make this technique suitable for widespread adoption in cardiovascular imaging analysis.
