Rapid Reconstruction of Four-dimensional MR Angiography of the Thoracic Aorta Using a Convolutional Neural Network. Haji-Valizadeh, H., Shen, D., Avery, R. J., Serhal, A. M., Schiffers, F. A., Katsaggelos, A. K., Cossairt, O. S., & Kim, D. Radiology: Cardiothoracic Imaging, 2(3):e190205, jun, 2020.
Rapid Reconstruction of Four-dimensional MR Angiography of the Thoracic Aorta Using a Convolutional Neural Network [link]Paper  doi  abstract   bibtex   
Purpose: To implement an integrated reconstruction pipeline including a graphics processing unit (GPU)–based convolutional neural network (CNN) architecture and test whether it reconstructs four-dimensional non-Cartesian, non–contrast material–enhanced MR angiographic k-space data faster than a central processing unit (CPU)–based compressed sensing (CS) reconstruction pipeline, without significant losses in data fidelity, summed visual score (SVS), or arterial vessel–diameter measurements. Materials and Methods: Raw k-space data of 24 patients (18 men and six women; mean age, 56.8 years 6 11.8 [standard deviation]) suspected of having thoracic aortic disease were used to evaluate the proposed reconstruction pipeline derived from an open-source threedimensional CNN. For training, 4800 zero-filled images and the corresponding CS-reconstructed images from 10 patients were used as input-output pairs. For testing, 6720 zero-filled images from 14 different patients were used as inputs to a trained CNN. Metrics for evaluating the agreement between the CNN and CS images included reconstruction times, structural similarity index (SSIM) and normalized root-mean-square error (NRMSE), SVS (3 = nondiagnostic, 9 = clinically acceptable, 15 = excellent), and vessel diameters. Results: The mean reconstruction time was 65 times and 69 times shorter for the CPU-based and GPU-based CNN pipelines (216.6 seconds 6 40.5 and 204.9 seconds 6 40.5), respectively, than for CS (14 152.3 seconds 6 1708.6) (P, .001). Compared with CS as practical ground truth, CNNs produced high data fidelity (SSIM = 0.94 6 0.02, NRMSE = 2.8% 6 0.4) and not significantly different (P = .25) SVS and aortic diameters, except at one out of seven locations, where the percentage difference was only 3% (ie, clinically irrelevant). Conclusion: The proposed integrated reconstruction pipeline including a CNN architecture is capable of rapidly reconstructing timeresolved volumetric cardiovascular MRI k-space data, without a significant loss in data quality, thereby supporting clinical translation of said non–contrast-enhanced MR angiograms.
@article{Hassan2020,
abstract = {Purpose: To implement an integrated reconstruction pipeline including a graphics processing unit (GPU)–based convolutional neural network (CNN) architecture and test whether it reconstructs four-dimensional non-Cartesian, non–contrast material–enhanced MR angiographic k-space data faster than a central processing unit (CPU)–based compressed sensing (CS) reconstruction pipeline, without significant losses in data fidelity, summed visual score (SVS), or arterial vessel–diameter measurements. Materials and Methods: Raw k-space data of 24 patients (18 men and six women; mean age, 56.8 years 6 11.8 [standard deviation]) suspected of having thoracic aortic disease were used to evaluate the proposed reconstruction pipeline derived from an open-source threedimensional CNN. For training, 4800 zero-filled images and the corresponding CS-reconstructed images from 10 patients were used as input-output pairs. For testing, 6720 zero-filled images from 14 different patients were used as inputs to a trained CNN. Metrics for evaluating the agreement between the CNN and CS images included reconstruction times, structural similarity index (SSIM) and normalized root-mean-square error (NRMSE), SVS (3 = nondiagnostic, 9 = clinically acceptable, 15 = excellent), and vessel diameters. Results: The mean reconstruction time was 65 times and 69 times shorter for the CPU-based and GPU-based CNN pipelines (216.6 seconds 6 40.5 and 204.9 seconds 6 40.5), respectively, than for CS (14 152.3 seconds 6 1708.6) (P, .001). Compared with CS as practical ground truth, CNNs produced high data fidelity (SSIM = 0.94 6 0.02, NRMSE = 2.8% 6 0.4) and not significantly different (P = .25) SVS and aortic diameters, except at one out of seven locations, where the percentage difference was only 3% (ie, clinically irrelevant). Conclusion: The proposed integrated reconstruction pipeline including a CNN architecture is capable of rapidly reconstructing timeresolved volumetric cardiovascular MRI k-space data, without a significant loss in data quality, thereby supporting clinical translation of said non–contrast-enhanced MR angiograms.},
author = {Haji-Valizadeh, Hassan and Shen, Daming and Avery, Ryan J. and Serhal, Ali M. and Schiffers, Florian A. and Katsaggelos, Aggelos K. and Cossairt, Oliver S. and Kim, Daniel},
doi = {10.1148/ryct.2020190205},
issn = {2638-6135},
journal = {Radiology: Cardiothoracic Imaging},
month = {jun},
number = {3},
pages = {e190205},
title = {{Rapid Reconstruction of Four-dimensional MR Angiography of the Thoracic Aorta Using a Convolutional Neural Network}},
url = {http://pubs.rsna.org/doi/10.1148/ryct.2020190205},
volume = {2},
year = {2020}
}
Downloads: 0
About
Documentation
Keywords
Search
Users
Terms and Conditions of Use
Privacy Policy
Sitemap
© BibBase.org 2021