TY - GEN
T1 - Fast Dynamic Perfusion and Angiography Reconstruction Using an End-to-End 3D Convolutional Neural Network
AU - Yousefi, Sahar
AU - Hirschler, Lydiane
AU - van der Plas, Merlijn
AU - Elmahdy, Mohamed S.
AU - Sokooti, Hessam
AU - van Osch, Matthias J P
AU - Staring, Marius
PY - 2019
Y1 - 2019
N2 - Hadamard time-encoded pseudo-continuous arterial spin labeling (te-pCASL) is a signal-to-noise ratio (SNR)-efficient MRI technique for acquiring dynamic pCASL signals that encodes the temporal information into the labeling according to a Hadamard matrix. In the decoding step, the contribution of each sub-bolus can be isolated resulting in dynamic perfusion scans. When acquiring te-ASL both with and without flow-crushing, the ASL-signal in the arteries can be isolated resulting in 4D-angiographic information. However, obtaining multi-timepoint perfusion and angiographic data requires two acquisitions. In this study, we propose a 3D Dense-Unet convolutional neural network with a multi-level loss function for reconstructing multi-timepoint perfusion and angiographic information from an interleaved 50 % -sampled crushed and 50 % -sampled non-crushed data, thereby negating the additional scan time. We present a framework to generate dynamic pCASL training and validation data, based on models of the intravascular and extravascular te-pCASL signals. The proposed network achieved SSIM values of 97.3 ± 1.1 and 96.2 ± 11.1 respectively for 4D perfusion and angiographic data reconstruction for 313 test data-sets.
AB - Hadamard time-encoded pseudo-continuous arterial spin labeling (te-pCASL) is a signal-to-noise ratio (SNR)-efficient MRI technique for acquiring dynamic pCASL signals that encodes the temporal information into the labeling according to a Hadamard matrix. In the decoding step, the contribution of each sub-bolus can be isolated resulting in dynamic perfusion scans. When acquiring te-ASL both with and without flow-crushing, the ASL-signal in the arteries can be isolated resulting in 4D-angiographic information. However, obtaining multi-timepoint perfusion and angiographic data requires two acquisitions. In this study, we propose a 3D Dense-Unet convolutional neural network with a multi-level loss function for reconstructing multi-timepoint perfusion and angiographic information from an interleaved 50 % -sampled crushed and 50 % -sampled non-crushed data, thereby negating the additional scan time. We present a framework to generate dynamic pCASL training and validation data, based on models of the intravascular and extravascular te-pCASL signals. The proposed network achieved SSIM values of 97.3 ± 1.1 and 96.2 ± 11.1 respectively for 4D perfusion and angiographic data reconstruction for 313 test data-sets.
KW - 4D magnetic resonance angiography (MRA)
KW - 4D perfusion
KW - Convolutional neural network (CNN)
KW - Hadamard time-encoded ASL
KW - MRI reconstruction
KW - Pseudo-continuous arterial spin labeling (pCASL)
UR - http://www.scopus.com/inward/record.url?scp=85076232200&partnerID=8YFLogxK
U2 - 10.1007/978-3-030-33843-5_3
DO - 10.1007/978-3-030-33843-5_3
M3 - Conference contribution
AN - SCOPUS:85076232200
SN - 978-3-030-33842-8
VL - 11905
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 25
EP - 35
BT - Machine Learning for Medical Image Reconstruction
A2 - Knoll, Florian
A2 - Maier, Andreas
A2 - Rueckert, Daniel
A2 - Ye, Jong Chul
PB - Springer
CY - Cham
T2 - 2nd International Workshop on Machine Learning for Medical Image Reconstruction, MLMIR 2019 held in Conjunction with 22nd International Conference on Medical Image Computing and Computer-Assisted Intervention, MICCAI 2019
Y2 - 17 October 2019 through 17 October 2019
ER -