Keywords: Vascular, Blood vessels, Super-Resolution

Motivation: The Fourier transform (FT) reconstruction has convenient implementation and stable performance; however, it has the problem of poor resolving power.

Goal(s): Our goal is to bypass the Fourier transform to obtain MR images, thereby solving the problem of poor resolution and achieving super-resolution imaging.

Approach: We were inspired by array signal processing theory and proposed an approach based on the Multiple Signal Classification (MUSIC) algorithm called MUSIC-MRI.

Results: Our phantom experiments suggest that the resolution ability of MUSIC-MRI is approximately 2x2 better than that of the 2D Fourier transform. Our in-vivo vascular imaging experiments show that the MUSIC-MRI significantly promotes the actual resolution.

Impact: MUSIC-MRI can break through the Rayleigh Limit of Fourier transform and significantly increase the actual resolution ability of the reconstructed images. Scientists or clinicians may use MUSIC-MRI to image very small structures and lesions without modifying MR sequences.

Keywords: Image Reconstruction, Quantitative Susceptibility mapping, EPI, distortion correction

Motivation: MR susceptibility mapping serves as a highly valuable tool in various neuroscientific and clinical applications.

Goal(s): This innovative approach is designed to facilitate fast and robust high-resolution whole-brain imaging and quantitative susceptibility mapping (QSM).

Approach: In this study, our primary objective was to create a distortion-free 3D-EPI with blip-up/down acquisition (BUDA), incorporating controlled aliasing in parallel imaging (CAIPI) sampling, and applying artificial sparsity enhanced deep learning image reconstruction.

Results: Our developed technique holds the potential to produce distortion-free high-resolution whole-brain quantitative susceptibility mapping in just 12s at 3T and 9s at 7T, achieving an impressive resolution of 1 mm isotropic.

Impact: The proposed 3D-BUDA, incorporating a 2D CAIPIRINHA acquisition sequence with artificial sparsity-enhanced self-supervised deep learning reconstruction, demonstrated its ability to deliver rapid, distortion-free, high-resolution, whole-brain T2*-weighted imaging and QSM.

Keywords: Image Reconstruction, Quantitative Imaging

Motivation: Routine clinical images are a massive data source for machine learning. The previously introduced e-CAMP method can convert T₂-weighted images of clinical TSE acquisition to quantitative T₂ maps, but it requires tuning of many parameters, impeding widespread implementation.

Goal(s): To present an algorithm that requires few parameter choices, is robust to those parameter values, and is faster to convergence. 

Approach: Projected Gradient Descent ensures efficient enforcement of the T₂-decay model constraint and greatly eliminates parameter tuning. e-CAMP is further enhanced by phase conjugacy with Virtual Conjugate Coils.

Results: The efficient and robust implementation of e-CAMP shows accurate T₂ map reconstruction.

Impact: Rather than acquire specific yet time-consuming quantitative images, e-CAMP can efficiently standardize the existing qualitative images from routine clinical scans and exploit the enormous amount of images to create dataset for large-scale machine learning.

Keywords: Image Reconstruction, Spectroscopy, Brain, High-Field MR, Image Reconstruction

Motivation: Magnetic resonance spectroscopic imaging (MRSI) is a unique method for non-invasive mapping of brain neurochemistry. While the latest advancements in acquisition enable whole-brain high-resolution metabolic imaging, these methods have lengthy reconstruction times that limit the clinical use.

Goal(s):  To realize a fast end-to-end reconstruction pipeline for high-resolution whole-brain MRSI compatible with online processing and clinical use. 

Approach: We developed a rapid deep-learning reconstruction pipeline for 3D non-Cartesian Compressed-Sensing MRSI (ECCENTRIC).

Results: Our approach reconstructs in a few minutes high-resolution ECCENTRIC (k,t) data. We demonstrate a 60-fold speed-up in reconstruction time, facilitating the use in clinical routine.

Impact: We present Deep-ECCENTRIC: a deep-learning pipeline for end-to-end reconstruction of 3D non-Cartesian Compressed-Sensing MRSI. We showcase spatially precise reconstructions with high spectral consistency, at a 60-fold speed-up over conventional reconstructions, which facilitates the clinical use of fast high-resolution MRSI.

Keywords: Image Reconstruction, Brain

Motivation: The fetal brain MRI 3D volume is critical for development assessment. However, the inevitable fetal motion during MRI acquisition makes it challenging to reconstruct a high-quality fetal brain 3D volume from multiple stacks.

Goal(s):  Herein, we propose a novel deep learning method for automated fetal brain MRI 3D volume reconstruction.

Approach: Firstly, a multi-scale feature fusion model is proposed to solve arbitrary motion correction. Secondly, an initial 3D volume is estimated by point spread function. Next, the proposed residual-based model is used to improve the quality of the initial 3D volume.

Results: The results demonstrate that the proposed method is effective and efficient.

Impact: The proposed end-to-end method based on deep learning can solve arbitrary motion correction of 2D slices and reconstruct high-resolution fetal brain MRI 3D volumes effectively and efficiently.

Keywords: Image Reconstruction, Multi-Contrast

Motivation: The conventional multi-contrast 3D reconstruction process is time-consuming and lacks sufficient acceleration factors.

Goal(s): To achieve highly accelerated multi-contrast brain imaging while enhancing reconstruction efficiency, the learnable CNN network is utilized.

Approach: Deep learning regularized SNMs (Deep SNMs) is developed by unrolling parallel imaging reconstruction using spatial nulling maps (SNMs) with CNN regularization. The network is iteratively expanded by gradient descent blocks and 2D convolution blocks.

Results: Compared to L1 regularized SNMs, the learnable CNN regularization simplifies reconstruction complexity and attains higher image quality. DeepSNMs achieves multi-contrast volumetric brain imaging reconstruction under caipirinha 9x and 12x acceleration.

Impact: This work successfully accomplishes multi-contrast volumetric brain imaging with 9-fold and 12-fold caipirinha acceleration. By addressing the time-consuming challenge of reconstructing multi-contrast 3D images, this work effectively utilizes and integrates information from multiple contrasts concurrently.

Keywords: Image Reconstruction, Brain

Motivation: Inspired by DR-CAM-GAN's progress in CS-MRI, we embraced ESSGAN with self-attention mechanisms.

Goal(s): To assess CBAM's impact on ESSGAN's ability to enhance CS-MRI reconstruction across diverse sampling rates.

Approach: Implemented ESSGAN+CBAM and performed experiments using T1-weighted brain images from the MICCAI 2023 dataset. Ablation studies compared DR-CAM-GAN, ESSGAN, ESSGAN+CAM, and ESSGAN+CBAM across varying sampling rates.

Results: At a 10% low sampling rate, ESSGAN and ESSGAN+CBAM demonstrated similar performance. Nevertheless, at higher sampling rates (≥20%), ESSGAN+CBAM outperformed all other models, affirming its effectiveness across evaluation metrics.

Impact: The study reveals that the integration of CBAM modules significantly enhances ESSGAN's performance in CS-MRI, particularly at higher undersampling rates, making it a valuable tool for rapid and accurate image reconstruction in clinical settings.

Keywords: Image Reconstruction, Spectroscopy

Motivation: Hyperpolarized (HP) ¹³C magnetic resonance spectroscopic imaging (MRSI) is efficient and reliable in assessing the aggressiveness of tumors and their response to treatments.

Goal(s): To incorporate a deep learning prior with k-space data fidelity for accelerating HP ¹³C MRSI.

Approach: Singular maps were generated from synthetic phantom datasets simulated by two-site exchange models and used to train the deep learning prior, which was then employed with the fidelity term to reconstruct the undersampled k-space data.

Results: The proposed method was demonstrated feasibility and generalizability on varied synthetic cancer datasets, and showed improved accuracy in value and location of tumors compared to other methods.

Impact: The proposed model could be considered as a general framework that extended the application of deep learning to MRSI reconstruction, which could be applied in varied cancer datasets.

Keywords: Image Reconstruction, Low-Field MRI

Motivation: While the emerging ULF MRI shows potential of low-cost and point-of-care imaging applications, its image quality is poor and the scan time is long.

Goal(s): To reduce the ULF brain MRI scan time through deep learning image reconstruction from partial Fourier and uniformly undersampled data.

Approach: We proposed a DL reconstruction method for fast 3D brain MRI at 0.055T by applying the 3D DL image reconstruction to undersampled 3D k-space data, achieving speed up of 2x over our newly developed partial Fourier reconstruction and superresolution (PF-SR) method.

Results: Our preliminary results show the proposed method could reduce noise, artifacts, and enhance spatial resolution.

Impact: Our model can work with uniformly undersampled data, leading to acceleration factor of 2, and a PF sampling of at a fraction of 0.7. Our development enables fast and quality whole-brain MRI at 0.055T, indicating potential for widespread biomedical applications.

Keywords: Image Reconstruction, Image Reconstruction

Motivation: Diffusion models, the latest generative modeling approach, hold significant promise for MRI reconstruction tasks.

Goal(s): Due to the lengthy and slow diffusion inverse process, diffusion model are challenging to apply directly to MR reconstruction tasks.

Approach: We employ SwinrnNet for initial reconstruction of undersampled images and introduce the DDDC module to supplement details, obviating the need for a protracted full reverse diffusion process.

Results: Our method achieves high-quality reconstructions within 3 seconds, outperforming SOTA approaches in quantitative metrics. It exhibits superior reconstruction speed compared to other diffusion model methods, with a remarkable PSNR of 38.28 dB in the case of 5x acceleration.

Impact: The DDDC module we propose effectively enhances reconstruction quality and can be applied extensively to any reconstruction model proposed by other researchers. With only a minimal time investment, it significantly improves image quality.

Keywords: Image Reconstruction, Artifacts, Single-shot fast spin-echo; PROPELLER; Follicle number per ovary; Ovarian volume

Motivation: Transvaginal ultrasonography (TVUS) often underestimates follicle count (FC) compared to MRI. The repeatability of FC and ovarian volume (OV) assessment was still affected by motion artifact on conventional T2-weighted fast spin echo images.

Goal(s): To propose a more reliable MRI technique in assessing the FC and OV.

Approach: High-resolution single-shot fast spin echo (SSFSE) sequence was used to accelerate the acquisition speed, and AIR^TM Recon DL was used to compensate for noise in this study.

Results: Contributing to the improved time resolution and reduced noise, SSFSE-DL demonstrated better repeatability in FC and OV assessment compared to the widely used motion-robust PROPELLER technique.

Impact: High-resolution SSFSE sequence with DL reconstruction can be a reliable imaging method in the assessment of  ovarian morphology. It has a potential in determining the threshold value of FC for PCOM identification in future studies.

Keywords: Machine Learning/Artificial Intelligence, Diffusion/other diffusion imaging techniques, Diffusion acquisition, distortion reduction, deep learning

Motivation: Diffusion-weighted imaging (DWI) is typically based on single-shot echo-planar imaging (EPI), which is prone to magnetic field inhomogeneities-induced artifacts, such as geometric distortion and blurring. The multi-shot diffusion sequence, DIADEM, employing a dual spin-warp (SW) and EPI phase-encoding strategies, can produce distortion-free images at the cost of extended scan times. 

Goal(s): We proposed a deep learning-based distortion correction method for conventional DWI, using DIADEM as reference. 

Approach: The 3D neural network was trained to learn the mapping between the projections of the point-spread-function, PSF H(y,s) along the EPI phase-encoding (y) direction and the PSF-encoding (s) direction, respectively.

Results: It demonstrated reduced geometric distortion.

Impact: Conventional DWI sequence suffers from distortion caused by susceptibility. We proposed a deep learning-based distortion correction method, leveraging distortion-free DIADEM images as reference. Our method was demonstrated to reduce geometric distortion and imaging blurring without distortion calibration. 

Keywords: Image Reconstruction, Image Reconstruction, deep unrolling network, dynamic MRI, low-rank, sparse

Motivation: The unrolling networks that combine low-rank (LR) and sparse priors have the potential to enhance the reconstruction performance. However, the underlying iterative algorithm for solving the model of joint LR and sparse constraint is complicated, resulting in redundant network structure.

Goal(s): To propose a simple yet effective alternating unrolling framework that exploits jointly LR and sparse prior for robust reconstruction of highly accelerated dynamic MRI data.

Approach: Instead of strictly unrolling the iterative algorithm, we propose a novel "DC-LowRank-DC-Sparse" alternating framework. 

Results: The proposed network (AlteRS-Net) outperforms the SOTA unrolling networks regarding both visualization and quantitative evaluation of PSNR and SSIM.

Impact: A novel alternating framework for unrolling network of jointly low-rank and sparse prior is established for accelerating dynamic MRI. This alternating concept could serve as an inspiration for the design of unrolling networks that combine other priors in various applications.

Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, Strain

Motivation: Accelerated cardiac cine MRI is prone to motion artifacts and underestimation of imaging biomarkers. Furthermore, many approaches lack prospective evaluation.

Goal(s): Improve data-driven reconstruction of cardiac cine MRI and enable inline reconstruction with improved estimation of strain parameters.

Approach: Training a neural network based on a Variational Network combined with intermediate conjugate gradient optimizations and evaluation on retrospectively undersampled data. Inline integration into scanner software using the FIRE framework and prospective evaluation in terms of image quality and cardiac strain parameters.

Results: The proposed network outperformed established compressed sensing approaches both retrospectively and prospectively, and in both image quality and cardiac strain estimation.

Impact: Our research enables inline reconstruction of highly accelerated (up to real-time) cardiac cine MRI with high motion fidelity and improved strain estimation compared to well-established compressed sensing approaches.

Keywords: Machine Learning/Artificial Intelligence, Image Reconstruction, Super Resolution, BLADE sequence

Motivation: Conventional super-resolution techniques are not applicable to magnetic resonance images reconstructed from BLADE sequences, where four corners of the k-space are missing.

Goal(s): When BLADE data are fed into a common super-resolution model targeting ordinary Cartesian k-space data, strong Gibbs rings appear due to the truncation of  k-space. On this basis, we propose a deep learning-based method specifically for the super-resolution task with BLADE data.

Approach: We mainly used the Residual Density Network (RDN) and designed the downsampling method based on the characteristics of BLADE data.

Results: Experimental results show that our model is able to predict high-resolution MR images with fewer artifacts.

Impact: By applying our RDN-based model specifically adapted to BLADE data, the image matrix size can be increased by a factor of 2 to produce sharper images without increasing acquisition time.

Keywords: Machine Learning/Artificial Intelligence, Machine Learning/Artificial Intelligence, MPI, FPGA, Deep Learning, Reconstruction

Motivation: When using complex prior parameters for regularization in the MPI reconstruction method based on the system matrix, the process is very time-consuming and the preprocessing process is complex.

Goal(s): The purpose of this study is to simplify the reconstruction process and achieve efficient real-time reconstruction of magnetic particle images. 

Approach: Therefore, this study uses CNN neural network for reconstruction on FPGA, and tests the neural network reconstruction effect through simulation and customized MPI system(Fig. 1). 

Results: The results show that it is feasible to use CNN neural network for reconstruction on FPGA, achieving high efficiency and real-time performance of magnetic particle image reconstruction.

Impact: FPGA-based CNN reconstruction network can make desktop magnetic particle imaging easier and more efficient.