An end-to-end deep learning algorithm for motion-corrected reconstruction reduced reconstruction time to 24 seconds versus 2.5 hours with HD-PROST, with small T1 and T2 biases of -6.35 ms and -1.8 ms.
Does an end-to-end deep learning algorithm improve reconstruction time and maintain image quality compared to HD-PROST for 3D whole-heart joint T1/T2 mapping?
A novel deep learning-based reconstruction algorithm for 3D whole-heart T1/T2 mapping dramatically reduces reconstruction time to 24 seconds while maintaining quantitative accuracy.
To accelerate 3D whole-heart joint T 1 /T 2 mapping for myocardial tissue characterization using an end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data. A free-breathing high-resolution motion-compensated 3D joint T 1 /T 2 water/fat sequence is employed. The sequence consists of the acquisition of four interleaved volumes with 2-echo encoding, resulting in eight volumes with different contrasts. An end-to-end non-rigid motion-corrected reconstruction network is used to estimate high quality motion-corrected reconstructions from the eight multi-contrast undersampled data for subsequent joint T 1 /T 2 mapping. Reconstruction with the proposed approach was compared against state-of-the-art motion-corrected HD-PROST reconstruction. The proposed approach yields images with good visual agreement compared to the reference reconstructions. The comparison of the quantitative values in the T 1 and T 2 maps showed the absence of systematic errors, and a small bias of − 6.35 ms and − 1.8 ms, respectively. The proposed reconstruction time was 24 seconds in comparison to 2.5 hours with motion-corrected HD-PROST, resulting in a reconstruction speed-up of over 370 times. In conclusion, this study presents a promising method for efficient whole-heart myocardial tissue characterization. Specifically, the research highlights the potential of the multi-contrast end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data. The findings underscore its ability to compute T 1 and T 2 values with good agreement when compared to the reference motion-corrected HD-PROST method, while substantially reducing reconstruction time.
Felsner et al. (Mon,) conducted a other in Myocardial tissue characterization. End-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction vs. Motion-corrected HD-PROST reconstruction was evaluated on Reconstruction time and quantitative T1/T2 map values. An end-to-end deep learning algorithm for motion-corrected reconstruction reduced reconstruction time to 24 seconds versus 2.5 hours with HD-PROST, with small T1 and T2 biases of -6.35 ms and -1.8 ms.