scholarly journals Coronary Centerline Extraction from CCTA Using 3D-UNet

2021 ◽  
Vol 13 (4) ◽  
pp. 101
Author(s):  
Alexandru Dorobanțiu ◽  
Valentin Ogrean ◽  
Remus Brad

The mesh-type coronary model, obtained from three-dimensional reconstruction using the sequence of images produced by computed tomography (CT), can be used to obtain useful diagnostic information, such as extracting the projection of the lumen (planar development along an artery). In this paper, we have focused on automated coronary centerline extraction from cardiac computed tomography angiography (CCTA) proposing a 3D version of U-Net architecture, trained with a novel loss function and with augmented patches. We have obtained promising results for accuracy (between 90–95%) and overlap (between 90–94%) with various network training configurations on the data from the Rotterdam Coronary Artery Centerline Extraction benchmark. We have also demonstrated the ability of the proposed network to learn despite the huge class imbalance and sparse annotation present in the training data.

2020 ◽  
Vol 16 (1) ◽  
pp. 155014771989595
Author(s):  
Huang Hao Lei ◽  
Li Hua Kang ◽  
Song Zhi Yuan ◽  
Wan Chen ◽  
Yao Qing

Objective: The objective was to evaluate the application of transesophageal echocardiography, left atrial appendage angiography, cardiac computed tomography angiography, and three-dimensional reconstruction model in the left atrial appendage occlusion. Methods: A total of 53 patients with persistent atrial fibrillation hospitalized from January 2015 to February 2019 were selected. Transesophageal echocardiography and cardiac computed tomography angiography were performed simultaneously before operation, and three-dimensional reconstruction of the left atrial appendage model was performed based on the cardiac computed tomography angiography findings. The morphology and size of the left atrial appendage were displayed by the left atrial appendage angiography during the operation. Results: Cardiac computed tomography angiography revealed filling defects in the left atrial appendage in four patients, whereas transesophageal echocardiography revealed thrombosis in two patients of the four patients. According to the cardiac computed tomography angiography results, the morphology of the left atrial appendage was classified into chicken wing–like, cauliflower-like, wind vane–like, and cactus-like. The diameters and depths of the left atrial appendage measured by cardiac computed tomography angiography and three-dimensional reconstruction model were found to be the greatest, and the orifice sizes and depths of the left atrial appendage measured by the left atrial appendage angiography were the smallest. Conclusion: Transesophageal echocardiography, left atrial appendage angiography, cardiac computed tomography angiography, and three-dimensional reconstruction model can evaluate the morphology and size of the left atrial appendage. The cardiac computed tomography angiography and three-dimensional reconstruction model could demonstrate the internal structure of the left atrial appendage more clearly.


2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Michael Hammer ◽  
Muhtashim Mian ◽  
Levi Elhadad ◽  
Mary Li ◽  
Idan Roifman

Abstract Background Appropriate use criteria (AUC) have been developed in response to growth in cardiac imaging utilization and concern regarding associated costs. Cardiac computed tomography angiography (CCTA) has emerged as an important modality in the evaluation of coronary artery disease, however its appropriate utilization in actual practice is uncertain. Our objective was to determine the appropriate utilization of CCTA in a large quaternary care institution and to compare appropriate utilization pre and post publication of the 2013 AUC guidelines. We hypothesized that the proportion of appropriate CCTA utilization will be similar to those of other comparable cardiac imaging modalities and that there would be a significant increase in appropriate use post AUC publication. Methods We employed a retrospective cohort study design of 2577 consecutive patients undergoing CCTA between January 1, 2012 and December 30, 2016. An appropriateness category was assigned for each CCTA. Appropriateness classifications were compared pre- and post- AUC publication via the chi-square test. Results Overall, 83.5% of CCTAs were deemed to be appropriate based on the AUC. Before the AUC publication, 75.0% of CCTAs were classified as appropriate whereas after the AUC publication, 88.0% were classified as appropriate (p < 0.001). The increase in appropriate utilization, when extrapolated to the Medicare population of the United States, was associated with potential cost savings of approximately $57 million per year. Conclusions We report a high rate of appropriate use of CCTA and a significant increase in the proportion of CCTAs classified as appropriate after the AUC publication.


2017 ◽  
Vol 23 (10) ◽  
pp. S80
Author(s):  
Moeko Suzuki ◽  
Teruyoshi Uetani ◽  
Jun Aono ◽  
Takayuki Nagai ◽  
Kazuhisa Nishimura ◽  
...  

1997 ◽  
Vol 82 (3) ◽  
pp. 998-1002 ◽  
Author(s):  
Nicolas Pettiaux ◽  
Marie Cassart ◽  
Manuel Paiva ◽  
Marc Estenne

Pettiaux, Nicolas, Marie Cassart, Manuel Paiva, and Marc Estenne. Three-dimensional reconstruction of human diaphragm with the use of spiral computed tomography. J. Appl. Physiol. 82(3): 998–1002, 1997.—We developed a technique of diaphragm imaging by using spiral computed tomography, and we studied four normal subjects who had been previously investigated with magnetic resonance imaging (A. P. Gauthier, S. Verbanck, M. Estenne, C. Segebarth, P. T. Macklem, and M. Paiva. J. Appl. Physiol. 76: 495–506, 1994). One acquisition of 15- to 25-s duration was performed at residual volume, functional residual capacity, functional residual capacity plus one-half inspiratory capacity, and total lung capacity with the subject holding his breath and relaxing. From these acquisitions, 20 coronal and 30 sagittal images were reconstructed at each lung volume; on each image, diaphragm contour in the zone of apposition and in the dome was digitized with the software Osiris, and the digitized silhouettes were used for three-dimensional reconstruction with Matlab. Values of length and surface area for the diaphragm, the dome, and the zone of apposition were very similar to those obtained with magnetic resonance imaging. We conclude that satisfactory three-dimensional reconstruction of the in vivo diaphragm may be obtained with spiral computed tomography, allowing accurate measurements of muscle length, surface area, and shape.


2021 ◽  
pp. 021849232110609
Author(s):  
Niraj Nirmal Pandey ◽  
Mumun Sinha ◽  
Arun Sharma ◽  
Ambuj Roy

A 25-year-old, previously asymptomatic female, presented to the outpatient clinic with episodic palpitations for past 6 months. She was acyanotic and showed no peripheral stigmata of infective endocarditis. Transthoracic echocardiography revealed dilated right ventricle with severe low-pressure pulmonary regurgitation. A cardiac computed tomography angiography performed for evaluation of pulmonary arterial circulation and intracardiac anatomy revealed isolated absence of posterior pulmonary cusp. The two other (right anterior and left anterior) cusps were normal and covered only part of the valve orifice, resulting in pulmonary insufficiency. The main pulmonary artery showed asymmetric dilatation. No other structural heart defects were noted.


Sign in / Sign up

Export Citation Format

Share Document