An introduction to an algorithm of Maximum-Gradient for extraction of biomechanical behavior of Renal and abdominal Aorta arterial walls with consecutive Ultrasound images


Authors:

Sima Ghorbani (MS.c), Department of Medical Physics, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran. Manijhe Mokhtari-Dizaji (Ph.D), Department of Medical Physics, Faculty of Medical Sciences, Tarbiat Modares University, Tehran, Iran.

Correspondence:

mokhtarm@modares.ac.ir

Aim: In the present study, we used a new computerized analytical method using maximum-gradient algorithms for extraction of instantaneous changes of renal and abdominal aorta arterial walls and studying their biomechanical behavior with sequential ultrasound images.

Methods: This algorithm was examined for evaluation of renal and abdominal aorta arteries behavior in one healthy person. Boundary conditions and limitations, arteries diameter and thickness were extracted from consecutive ultrasound images for three cardiac cycles. To validate the automated analysis, we performed a manual measurement and compared it with automated method, and then diameter of both arteries were extracted with automated analysis, over three cardiac cycles and finally compared with manual measurements.

Results: There was a significant correlation between automatic and manual measurements (correlation coefficient of 0.750 for abdominal aorta and 0.873 for left renal artery). This algorithm confirmed a good agreement between two mentioned methods, automated analysis and manual measurements.

Conclusions: According to the results of this research, computerized analytical method automatically can extract instantaneous changes, diameter and thickness of arterial wall in consecutive ultrasound images. Therefore we can predict these diseases which are related to arterial dysfunction such as renal failure, chronic kidney disease, renal artery stenosis, hypertension and other kind of diseases before any destruction in the artery or organ.

Keywords: Ultrasound, Biomechanical Behavior, Maximum-Gradient Algorithms.