Deep learning–enhanced T₁ mapping with spatial‐temporal and physical constraint

Abstract

Purpose To propose a reconstruction framework to generate accurate T1 maps for a fast MR T1 mapping sequence. Methods A deep learning–enhanced T1 mapping method with spatial‐temporal and physical constraint (DAINTY) was proposed. This method explicitly imposed low‐rank and sparsity constraints on the multiframe T1‐weighted images to exploit the spatial‐temporal correlation. A deep neural network was used to efficiently perform T1 mapping as well as denoise and reduce undersampling artifacts. Additionally, the physical constraint was used to build a bridge between low‐rank and sparsity constraint and deep learning prior, so the benefits of constrained reconstruction and deep learning can be both available. The DAINTY method was trained on simulated brain data sets, but tested on real acquired phantom, 6 healthy volunteers, and 7 atherosclerosis patients, compared with the narrow‐band k‐space‐weighted image contrast filter conjugate‐gradient SENSE (NK‐CS) method, kt‐sparse‐SENSE (kt‐SS) method, and low‐rank plus sparsity (L+S) method with least‐squares T1 fitting and direct deep learning mapping. Results The DAINTY method can generate more accurate T1 maps and higher‐quality T1‐weighted images compared with other methods. For atherosclerosis patients, the intraplaque hemorrhage can be successfully detected. The computation speed of DAINTY was 10 times faster than traditional methods. Meanwhile, DAINTY can reconstruct images with comparable quality using only 50% of k‐space data. Conclusion The proposed method can provide accurate T1 maps and good‐quality T1‐weighted images with high efficiency.

Motion correction for native myocardial T1 mapping using self‐supervised deep learning registration with contrast separation

“…In this study, a DenseAttention UNet from our previous study 34 was used to perform the deformation field prediction. The architecture of the network is shown in Supporting Figure S2.…”

Deep MR parametric imaging with the learned L + S model and attention mechanism

“…In theory, the $$T1\rho$$‐weighted images share an approximate structure except the signal intensity. As proposed in DAINTY [25] method, the L + S decomposition method has been applied in T1 imaging, which inspired us to introduce this method to other parametric imaging processes. Therefore, consider applying L + S ‐Net to the reconstruction part of $$T1\rho$$ imaging, that is, to reconstruct multiple under‐sampled (usually 5) $$T1\rho$$ weighted k ‐space data of different contrasts into clear $$T1\rho$$‐weighted images.…”

Integrating mechanism-based modeling with biomedical imaging to build practical digital twins for clinical oncology

“…For example, Cohen et al developed a DNN applied to the MR fingerprinting technique, which achieved a reconstruction of high spatial resolution T1 and T2 maps 300 and 5000 times faster, respectively, than the conventional dictionary-matching methods [193]. This increase in resolution provides access to a more refined domain on which to perform calculations, more accurate segmentation of important anatomical features [194], and more refined parametric maps of tissue properties [195].…”