Refine
Has Fulltext
- yes (3)
Is part of the Bibliography
- yes (3)
Document Type
- Journal article (2)
- Preprint (1)
Language
- English (3)
Keywords
- 3 T (1)
- CMR (1)
- Deep learning (1)
- Extracellular volume (1)
- MOLLI (1)
- Myocardial infarction (1)
- SASHA (1)
- Scar (1)
- Segmentation (1)
- ShMOLLI (1)
Institute
- Deutsches Zentrum für Herzinsuffizienz (DZHI) (3) (remove)
Purpose
To fully automatically derive quantitative parameters from late gadolinium enhancement (LGE) cardiac MR (CMR) in patients with myocardial infarction and to investigate if phase sensitive or magnitude reconstructions or a combination of both results in best segmentation accuracy.
Methods
In this retrospective single center study, a convolutional neural network with a U-Net architecture with a self-configuring framework (“nnU-net”) was trained for segmentation of left ventricular myocardium and infarct zone in LGE-CMR. A database of 170 examinations from 78 patients with history of myocardial infarction was assembled. Separate fitting of the model was performed, using phase sensitive inversion recovery, the magnitude reconstruction or both contrasts as input channels.
Manual labelling served as ground truth. In a subset of 10 patients, the performance of the trained models was evaluated and quantitatively compared by determination of the Sørensen-Dice similarity coefficient (DSC) and volumes of the infarct zone compared with the manual ground truth using Pearson’s r correlation and Bland-Altman analysis.
Results
The model achieved high similarity coefficients for myocardium and scar tissue. No significant difference was observed between using PSIR, magnitude reconstruction or both contrasts as input (PSIR and MAG; mean DSC: 0.83 ± 0.03 for myocardium and 0.72 ± 0.08 for scars). A strong correlation for volumes of infarct zone was observed between manual and model-based approach (r = 0.96), with a significant underestimation of the volumes obtained from the neural network.
Conclusion
The self-configuring nnU-net achieves predictions with strong agreement compared to manual segmentation, proving the potential as a promising tool to provide fully automatic quantitative evaluation of LGE-CMR.
Purpose
Image acquisition and subsequent manual analysis of cardiac cine MRI is time-consuming. The purpose of this study was to train and evaluate a 3D artificial neural network for semantic segmentation of radially undersampled cardiac MRI to accelerate both scan time and postprocessing.
Methods
A database of Cartesian short-axis MR images of the heart (148,500 images, 484 examinations) was assembled from an openly accessible database and radial undersampling was simulated. A 3D U-Net architecture was pretrained for segmentation of undersampled spatiotemporal cine MRI. Transfer learning was then performed using samples from a second database, comprising 108 non-Cartesian radial cine series of the midventricular myocardium to optimize the performance for authentic data. The performance was evaluated for different levels of undersampling by the Dice similarity coefficient (DSC) with respect to reference labels, as well as by deriving ventricular volumes and myocardial masses.
Results
Without transfer learning, the pretrained model performed moderately on true radial data [maximum number of projections tested, P = 196; DSC = 0.87 (left ventricle), DSC = 0.76 (myocardium), and DSC =0.64 (right ventricle)]. After transfer learning with authentic data, the predictions achieved human level even for high undersampling rates (P = 33, DSC = 0.95, 0.87, and 0.93) without significant difference compared with segmentations derived from fully sampled data.
Conclusion
A 3D U-Net architecture can be used for semantic segmentation of radially undersampled cine acquisitions, achieving a performance comparable with human experts in fully sampled data. This approach can jointly accelerate time-consuming cine image acquisition and cumbersome manual image analysis.
Background
T1 mapping sequences such as MOLLI, ShMOLLI and SASHA make use of different technical approaches, bearing strengths and weaknesses. It is well known that obtained T1 relaxation times differ between the sequence techniques as well as between different hardware. Yet, T1 quantification is a promising tool for myocardial tissue characterization, disregarding the absence of established reference values. The purpose of this study was to evaluate the feasibility of native and post-contrast T1 mapping methods as well as ECV maps and its diagnostic benefits in a clinical environment when scanning patients with various cardiac diseases at 3 T.
Methods
Native and post-contrast T1 mapping data acquired on a 3 T full-body scanner using the three pulse sequences 5(3)3 MOLLI, ShMOLLI and SASHA in 19 patients with clinical indication for contrast enhanced MRI were compared. We analyzed global and segmental T1 relaxation times as well as respective extracellular volumes and compared the emerged differences between the used pulse sequences.
Results
T1 times acquired with MOLLI and ShMOLLI exhibited systematic T1 deviation compared to SASHA. Myocardial MOLLI T1 times were 19% lower and ShMOLLI T1 times 25% lower compared to SASHA. Native blood T1 times from MOLLI were 13% lower than SASHA, while post-contrast MOLLI T1-times were only 5% lower. ECV values exhibited comparably biased estimation with MOLLI and ShMOLLI compared to SASHA in good agreement with results reported in literature. Pathology-suspect segments were clearly differentiated from remote myocardium with all three sequences.
Conclusion
Myocardial T1 mapping yields systematically biased pre- and post-contrast T1 times depending on the applied pulse sequence. Additionally calculating ECV attenuates this bias, making MOLLI, ShMOLLI and SASHA better comparable. Therefore, myocardial T1 mapping is a powerful clinical tool for classification of soft tissue abnormalities in spite of the absence of established reference values.