Key result
A 3D U-Net architecture with transfer learning achieved human-level semantic segmentation of highly undersampled radial cardiac MRI (DSC up to 0.95 for left ventricle).
Why the study?
Image acquisition and manual analysis of cardiac cine MRI are time-consuming, prompting evaluation of whether deep learning could accelerate both scan time and postprocessing.
Does a 3D U-Net architecture with transfer learning accurately segment radially undersampled cardiac MRI compared to fully sampled data?
Does a 3D U-Net architecture with transfer learning accurately segment radially undersampled cardiac MRI compared to fully sampled data?
A 3D U-Net architecture with transfer learning can accurately segment radially undersampled cardiac MRI, potentially accelerating scan time and manual image analysis.
May accelerate radial cardiac MRI workflows; extends transfer learning segmentation but leaves prospective validation open.
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.
No takes yet. Share an insight, caveat, or question.
Wech et al. (2021) studied Cardiac MRI segmentation (n=592). 3D U-Net architecture with transfer learning vs. Pretrained model without transfer learning and fully sampled data was evaluated on Dice similarity coefficient (DSC) for left ventricle, myocardium, and right ventricle. A 3D U-Net architecture with transfer learning achieved human-level semantic segmentation of highly undersampled radial cardiac MRI (DSC up to 0.95 for left ventricle).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: