Abstract: PO1329
Analysis of Vascular Distensibility Measured by Ultrasound Speckle Tracking
Session Information
- Vascular Access
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Dialysis
- 704 Dialysis: Vascular Access
Authors
- Weitzel, William, Veterans Affairs Ann Arbor Health System, Ann Arbor, Michigan, United States
- Rajaram, Nirmala, Veterans Affairs Ann Arbor Health System, Ann Arbor, Michigan, United States
- Yessayan, Lenar Tatios, University of Michigan, Ann Arbor, Michigan, United States
- Zheng, Yihao, Worcester Polytechnic Institute, Worcester, Massachusetts, United States
- Thelen, Brian, Veterans Affairs Ann Arbor Health System, Ann Arbor, Michigan, United States
- Morgan, Timothy, Veterans Affairs Ann Arbor Health System, Ann Arbor, Michigan, United States
- Hamilton, James, Veterans Affairs Ann Arbor Health System, Ann Arbor, Michigan, United States
- Funes-Lora, Miguel Angel, University of Michigan, Ann Arbor, Michigan, United States
Background
We have developed a novel open source ultrasound software program that measures vascular strain and distensibility using conventional ultrasound Digital Imaging and Communications in Medicine (DICOM) data for use in the dialysis vascular access setting. In this study, we evaluated the variation in measurement from operator point selection and physiologic beat to beat variation of the arterial wall.
Methods
Ten subjects scheduled for arteriovenous fistula (AVF) creation were enrolled in the study. Ultrasound scanning of the brachial/radial arteries was performed. Ten users were prompted to select two points of interest at the top and bottom of the arterial vessel wall in each of the ten subjects. These points were tracked using the Kanade-Lucas-Tomasi (KLT) tracking algorithm.
Results
Sub-millimeter resolution (less than 100 micron) measurements were obtained. We found variation point selection by the users for the ten cine loops to be up to 120 pixels for the top and up to 140 pixels for the bottom of the vessel wall. The range in measured variation attributable to user point selection was 5.79 to 47.29% and inter-cardiac (physiologic) variation was 6.41 to 17.68%.
Conclusion
Despite the low resolution of conventional DICOM images, we are able to measure sub-millimeter distensibility. In order to better understand the physiologic variation in vascular wall compliance, a formalized approach to point selection is needed. We are evaluating algorithms and statistical ensemble methods for use in studies to predict AVF maturation.
From image cine loop (A), a single frame (B) is used to select vessel wall edge points, for image tracking (C), to determine sub-millimeter resolution wall strain and distensiblity (D) showing beat to beat variation.
Funding
- Veterans Affairs Support