Premium
Highly accelerated, model‐free diffusion tensor MRI reconstruction using neural networks
Author(s) -
Aliotta Eric,
Nourzadeh Hamidreza,
Sanders Jason,
Muller Donald,
Ennis Daniel B.
Publication year - 2019
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1002/mp.13400
Subject(s) - diffusion mri , fractional anisotropy , nuclear medicine , receiver operating characteristic , effective diffusion coefficient , artificial neural network , computer science , magnetic resonance imaging , medicine , artificial intelligence , mathematics , radiology , machine learning
Purpose The purpose of this study was to develop a neural network that accurately performs diffusion tensor imaging ( DTI ) reconstruction from highly accelerated scans. Materials and Methods This retrospective study was conducted using data acquired between 2013 and 2018 and was approved by the local institutional review board. DTI acquired in healthy volunteers (N = 10) was used to train a neural network, DiffNet , to reconstruct fractional anisotropy ( FA ) and mean diffusivity ( MD ) maps from small subsets of acquired DTI data with between 3 and 20 diffusion‐encoding directions. FA and MD maps were then reconstructed in volunteers and in patients with glioblastoma multiforme ( GBM , N = 12) using both DiffNet and conventional reconstructions. Accuracy and precision were quantified in volunteer scans and compared between reconstructions. The accuracy of tumor delineation was compared between reconstructed patient data by evaluating agreement between DTI ‐derived tumor volumes and volumes defined by contrast‐enhanced T1‐weighted MRI . Comparisons were performed using areas under the receiver operating characteristic curves ( AUC ). Results DiffNet FA reconstructions were more accurate and precise compared with conventional reconstructions for all acceleration factors. DiffNet permitted reconstruction with only three diffusion‐encoding directions with significantly lower bias than the conventional method using six directions (0.01 ± 0.01 vs 0.06 ± 0.01, P < 0.001). While MD ‐based tumor delineation was not substantially different with DiffNet ( AUC range: 0.888–0.902), DiffNet FA had higher AUC than conventional reconstructions for fixed scan time and achieved similar performance with shorter scans (conventional, six directions: AUC = 0.926, DiffNet, three directions: AUC = 0.920). Conclusion DiffNet improved DTI reconstruction accuracy, precision, and tumor delineation performance in GBM while permitting reconstruction from only three diffusion‐encoding directions.&!#6;