Automatic detection of aorto-femoral vessel trajectory from whole-body computed tomography angiography data sets

Abstract

The average Dice similarity indexes between the segmentations of the automatic method and observer 1 for the left ilio-femoral artery, the right ilio-femoral artery and the aorta were 0.977 ± 0.030, 0.980 ± 0.019, 0.982 ± 0.016; the average Dice similarity indexes between the segmentations of the automatic method and observer 2 were 0.950 ± 0.040, 0.954 ± 0.031 and 0.965 ± 0.019, respectively. The interobserver variability resulted in a Dice similarity index of 0.954 ± 0.038, 0.952 ± 0.031 and 0.969 ± 0.018 for the left iliofemoral artery, the right ilio-femoral artery and the aorta. The average minimal luminal diameters (MLDs) of the ilio-femoral artery were 6.03 ± 1.48, 5.70 ± 1.43 and 5.52 ± 1.32 mm for the automatic method, observer 1 and observer 2 respectively. The MLDs of the aorta were 13.43 ± 2.54, 12.40 ± 2.93 and 12.08 ± 2.40 mm for the automatic method, observer 1 and observer 2 respectively. The automatic measurement overestimated the MLD slightly in the ilio-femoral artery at the average by 0.323 mm (SD = 0.49 mm, p < 0.001) compared to observer 1 and by 0.51 mm (SD = 0.71 mm, p < 0.001) compared to observer 2. The proposed segmentation approach can automatically provide reliable measurements of the entire arterial accessing route that can be used to support TAVR procedures. To the best of our knowledges, this approach is the first fully automatic segmentation method of the whole aorto-femoral vessel trajectory in CTA images.

A novel software tool for semi-automatic quantification of thoracic aorta dilatation on baseline and follow-up computed tomography angiography

“…The automatic thoracic aorta segmentation scheme was based on the centerline extraction and subsequent contour detection methods. The centerline extraction method was similar to the algorithm that we developed and described previously [ 9 ], based on the wave-propagation algorithm, Gaussian probabilistic distribution model and Dijkstra shortest path algorithm. The previous automatic landmark detection algorithm was modified from the previous algorithm which allows the detection of two femoral end points to the aortic root end point.…”

“…The aligned follow-up CTA image was processed by the centerline-based adaptive threshold method [ 12 ] to reduce the influence of the surrounding tissues in the background, such as high intensity tissue like bone, and low intensity tissue like muscle.…”

“…With the alignment of the baseline and follow-up images as described in the previous step, the position and shape of the aorta in the two images became the same with the exception of differences due to the dilatation. By using the aligned follow-up CTA image as the cost function image and the baseline aorta contour as the initial contour, the subdivision surface fitting algorithm will deform the initial contour to detect the edge with the highest gradient in the cost function image [ 12 ]. Following the segmentation of the aorta by subdivision fitting, a region-growing algorithm was used to detect in detail the aortic arch and its branches.…”

Cardiovascular imaging 2016 in the International Journal of Cardiovascular Imaging

“…Published studies examined quantitative semi-automated methods for assessment of coronary luminal stenosis severity [70] and fully-automated techniques for the extraction of the entire arterial access route from the femoral artery to the aortic root for TAVR evaluation [71] or the carotid arteries in CTA in the thorax and upper neck region [72]. Other studies examined subtraction CCTA in the evaluation of in-stent restenosis [73] and fusion of 3D echocardiography (3DE) with multidetector computed tomography (MDCT) to correlate territorial longitudinal strain (LS) with coronary stenosis [74].…”

Automatic aortic root segmentation in CTA whole-body dataset

“…With the centerline extracted automatically in whole-body CTA image in our previous research 15 , the location of aortic root was detected. With this location as the center, the region of interest (ROI) with the same size of the cardiac CTA image in world 3D coordinate was defined in whole-body CTA image.…”