Using the Rate of Change of Cross-Sectional Area Slices of the Left Ventricle as a Means for Quantifying Sphericity in 3D Cardiac Ultrasound Images
More info in the report and the demo video!
- demo of the automated software
- abstract
- background
- Methods (image processing, sphericity index ratio calculation, etc.)
- Automated slicing software
- Discussion
- Conclusion
As the population of the United States continues to grow older, the prevalence of heart disease increases. Ventricular remodeling is a result of a number of heart conditions where the left ventricle “balloons” outward as a result of compensatory mechanisms the heart must undergo to be able to pump blood efficiently to the rest of the body. 3D ultrasound technology has brought about novel ways to analyze the degree of dilation in a person’s left ventricle, but the volume of data that must be processed and the poorly defined ways of analyzing this data have resulted in a barrier towards using 3D ultrasound with greater frequency in the clinic.
In this project, we propose a novel way to determine the sphericity of the ventricle using 3D ultrasound data and a novel way to automate one of the previous methods that has been used to determine the sphericity of the ventricle using 2D ultrasound data. We believe that the use of these two methods combined will result in greater diagnostic power to render 3D ultrasound more useful towards saving the lives of patients in the clinic.
Image processing outputs for each step of the filtering process
Demonstration of the measurements to develop the elliptical area approximation of the ventricle
Screenshots of the automated slicing software
