English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

PURPOSE: To apply deep CNN to the segmentation task in myocardial arterialspin labeled (ASL) perfusion imaging and to develop methods that measureuncertainty and that adapt the CNN model to a specific false positive vs. falsenegative tradeoff.  METHODS: The Monte Carlo dropout (MCD) U-Net was trained on data from 22subjects and tested on data from 6 heart transplant recipients. Manualsegmentation and regional myocardial blood flow (MBF) were available forcomparison. We consider two global uncertainty measures, named Dice Uncertaintyand MCD Uncertainty, which were calculated with and without the use of manualsegmentation, respectively. Tversky loss function with a hyperparameter $\beta$was used to adapt the model to a specific false positive vs. false negativetradeoff.  RESULTS: The MCD U-Net achieved Dice coefficient of mean(std) = 0.91(0.04) onthe test set. MBF measured using automatic segmentations was highly correlatedto that measured using the manual segmentation ($R^2$ = 0.96). Dice Uncertaintyand MCD Uncertainty were in good agreement ($R^2$ = 0.64). As $\beta$increased, the false positive rate systematically decreased and false negativerate systematically increased.  CONCLUSION: We demonstrate the feasibility of deep CNN for automaticsegmentation of myocardial ASL, with good accuracy. We also introduce twosimple methods for assessing model uncertainty. Finally, we demonstrate theability to adapt the CNN model to a specific false positive vs. false negativetradeoff. These findings are directly relevant to automatic segmentation inquantitative cardiac MRI and are broadly applicable to automatic segmentationproblems in diagnostic imaging.

Accuracy, Uncertainty, and Adaptability of Automatic Myocardial ASL Segmentation using Deep CNN