Improving the accuracy of cardiac DTI by averaging the complex data

Scott, Andrew D.; Nielles‐Vallespin, Sonia; Ferreira, Pedro; McGill, Laura-Ann; Pennell, Dudley J.; Firmin, David

doi:10.1186/1532-429x-17-s1-o38

Cited by 1 publication

(1 citation statement)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Low baseline SNR thus degrades both accuracy and precision with LLS, but this effect does not appear to be present in DiffNet reconstructions. Similar noise‐induced biases are not generally observed for MD, with one possible exception being extreme cases where very low SNR baseline images are coupled with high b‐values and noise floor effects cause an apparent downward MD bias . This likely underlies the reason that MD accuracy was largely flat with LLS with varying N dir (Fig.…”

Section: Discussionsupporting

confidence: 54%

Highly accelerated, model‐free diffusion tensor MRI reconstruction using neural networks

et al. 2019

View full text Add to dashboard Cite

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 T1weighted 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 AE 0.01 vs 0.06 AE 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.

show abstract

Section: Discussionsupporting

confidence: 54%