Accuracy, uncertainty, and adaptability of automatic myocardial ASL segmentation using deep CNN

• Published in: Magnetic resonance in medicine Vol. 83; no. 5; pp. 1863 - 1874
• Main Authors: Do, Hung P., Guo, Yi, Yoon, Andrew J., Nayak, Krishna S.
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.05.2020
• Subjects: Adaptability; Artificial neural networks; automatic segmentation; Bayesian; Blood flow; Computer simulation; Convolution; Correlation analysis; deep convolutional neural network; Diagnostic systems; false‐positive and false‐negative tradeoff; Heart transplantation; Humans; Image processing; Image Processing, Computer-Assisted; Image segmentation; Magnetic Resonance Imaging; Measurement methods; Medical imaging; Monte Carlo dropout; Monte Carlo simulation; Myocardium; Neural networks; Neural Networks, Computer; Perfusion; Spin labeling; Tradeoffs; Uncertainty; uncertainty measure; Monte Carlo dropout; deep convolutional neural network; automatic segmentation; false-positive and false-negative tradeoff; uncertainty measure; Bayesian
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.28043

Purpose To apply deep convolution neural network to the segmentation task in myocardial arterial spin labeled perfusion imaging and to develop methods that measure uncertainty and that adapt the convolution neural network model to a specific false‐positive versus false‐negative tradeoff. Methods The Monte Carlo dropout U‐Net was trained on data from 22 subjects and tested on data from 6 heart transplant recipients. Manual segmentation and regional myocardial blood flow were available for comparison. We consider 2 global uncertainty measures, named “Dice uncertainty” and “Monte Carlo dropout uncertainty,” which were calculated with and without the use of manual segmentation, respectively. Tversky loss function with a hyperparameter β was used to adapt the model to a specific false‐positive versus false‐negative tradeoff. Results The Monte Carlo dropout U‐Net achieved a Dice coefficient of 0.91 ± 0.04 on the test set. Myocardial blood flow measured using automatic segmentations was highly correlated to that measured using the manual segmentation (R2 = 0.96). Dice uncertainty and Monte Carlo dropout uncertainty were in good agreement (R2 = 0.64). As β increased, the false‐positive rate systematically decreased and false‐negative rate systematically increased. Conclusion We demonstrate the feasibility of deep convolution neural network for automatic segmentation of myocardial arterial spin labeling, with good accuracy. We also introduce 2 simple methods for assessing model uncertainty. Finally, we demonstrate the ability to adapt the convolution neural network model to a specific false‐positive versus false‐negative tradeoff. These findings are directly relevant to automatic segmentation in quantitative cardiac MRI and are broadly applicable to automatic segmentation problems in diagnostic imaging.