Tavoos Rahmani-Cherati1, Manijhe Mokhtari-Dizaji2, Alireza Vajhi3, Abdorrazzagh Rostami3. 1. Department of Medical Physics, Faculty of Medical Sciences, Tarbiat Modares University, P.O. Box 14115-133, Tehran, Iran. 2. Department of Medical Physics, Faculty of Medical Sciences, Tarbiat Modares University, P.O. Box 14115-133, Tehran, Iran. mokhtarm@modares.ac.ir. 3. Department of Clinical Sciences, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran.
Abstract
PURPOSE: Many structural and dynamic properties of the arterial wall, e.g., lumen diameter and wall thickness, can be measured with non-invasive ultrasound techniques. We present a new computerized analysis method for measurement of instantaneous changes in far and near arterial walls in sequential ultrasound images. METHODS: In this method, two algorithms, i.e., maximum gradient and dynamic programming, were composed and implemented. Reference points and cost function were based on dynamic programming and maximum gradient, respectively. To evaluate this method, B-mode abdominal aorta ultrasound examinations were obtained for ten New Zealand White rabbits. Approximately 70 sequential ultrasound images spanning three cardiac cycles were analyzed from each examination to detect instantaneous changes in the far and near walls and lumen diameter of rabbit abdominal aorta. The maximum, minimum, and mean diameters extracted by the automated method were compared with the same parameters measured by manual tracing. RESULTS: There was no significant differences between the manual and automated methods according to paired t test analysis (p > 0.05). In a verification study, the correlation between the two methods was calculated (R (2) = 0.98, p < 0.05). Agreement between the automated and manual methods as determined by the Bland-Altman approach was excellent. CONCLUSION: It is concluded that the new computerized analysis method is a reliable technique to assess instantaneous changes in the rabbit arterial wall in sequential B-mode images. The variability between different laboratories will be reduced if the same analysis program is used. By using this method, not only was variation in the results of manual tracing by several observers eliminated, but the duration of image processing was also considerably reduced.
PURPOSE: Many structural and dynamic properties of the arterial wall, e.g., lumen diameter and wall thickness, can be measured with non-invasive ultrasound techniques. We present a new computerized analysis method for measurement of instantaneous changes in far and near arterial walls in sequential ultrasound images. METHODS: In this method, two algorithms, i.e., maximum gradient and dynamic programming, were composed and implemented. Reference points and cost function were based on dynamic programming and maximum gradient, respectively. To evaluate this method, B-mode abdominal aorta ultrasound examinations were obtained for ten New Zealand White rabbits. Approximately 70 sequential ultrasound images spanning three cardiac cycles were analyzed from each examination to detect instantaneous changes in the far and near walls and lumen diameter of rabbit abdominal aorta. The maximum, minimum, and mean diameters extracted by the automated method were compared with the same parameters measured by manual tracing. RESULTS: There was no significant differences between the manual and automated methods according to paired t test analysis (p > 0.05). In a verification study, the correlation between the two methods was calculated (R (2) = 0.98, p < 0.05). Agreement between the automated and manual methods as determined by the Bland-Altman approach was excellent. CONCLUSION: It is concluded that the new computerized analysis method is a reliable technique to assess instantaneous changes in the rabbit arterial wall in sequential B-mode images. The variability between different laboratories will be reduced if the same analysis program is used. By using this method, not only was variation in the results of manual tracing by several observers eliminated, but the duration of image processing was also considerably reduced.
Entities:
Keywords:
Arterial wall; Maximum gradient algorithm; Motion detection; Ultrasound