Improving deep PROPELLER MRI via synthetic blade augmentation and enhanced generalization.

In PROPELLER MRI, obtaining sufficient high-quality blade data remains a challenge, so the efficiency and generalization of deep learning-based reconstruction models are deteriorated. Due to narrow rotated and translated blades acquired in PROPELLER, the technique of data augmentation that is used for deep learning-based Cartesian MRI reconstruction cannot be directly applied. To address the issue, this paper introduces a novel approach for the generation of synthetic PROPELLER blades, and it is subsequently employed in data augmentation for undersampled blades reconstruction. The principal aim of this study is to address the challenges of reconstructing undersampled blades to enhance both image quality and computational efficiency. Evaluation metrics including PSNR, NMSE, and SSIM indicate superior performance of the model trained with augmented data compared to non-augmented counterparts. The synthetic blade augmentation significantly enhances the model's generalization capability and enables robust performance across varying imaging conditions. Furthermore, the study demonstrates the feasibility of utilizing synthetic blades exclusively in the training phase, suggesting a reduced dependency on real PROPELLER blades. This innovation in synthetic blade generation and data augmentation technique contributes to enhanced image quality and improved generalization capability of the associated deep learning model for PROPELLER MRI reconstruction.

Suppressing image blurring of PROPELLER MRI via untrained method.

Objective. Periodically rotated overlapping parallel lines with enhanced reconstruction (PROPELLER) used in magnetic resonance imaging (MRI) is inherently insensitive to motion artifacts but with an expense of around 60% increase in minimum scan time. An untrained deep learning method is proposed to accelerate PROPELLER MRI while suppressing image blurring.Approach. Several reconstruction methods have been developed to accelerate PROPELLER with reduced sampling on blades. However, image quality is degraded due to blurring. Deep learning has been applied to enhance MRI reconstruction quality, and external training data are therefore needed. In addition, the distribution shift problem in deep learning also exists between the external training data and to-be-reconstructed target blade data. This paper introduces an untrained neural network (UNN) to suppress image blurring, which is applied to improve PROPELLER MRI. This network structure was then incorporated into bladek-space.Results. The untrained method improved the blade image quality from brain MRI data. Furthermore, it enhanced the sharpness of the reconstructed image compared to PROPELLER reconstructions using parallel imaging methods and supervised learning methods using external training data. PROPELLER blade acquisition was accelerated by undersampling data with reduction factors 2, 3 and 4.Significance. The reported UNN enhanced PROPELLER method can improve image quality by suppressing blurring. External training data are not needed to mitigate the challenge of collecting high-quality clinical data for training without affecting clinical workflow and the standard care for patients.

Virtual Conjugate Coil for Improving KerNL Reconstruction.

Ker NL is a general kernel-based framework for auto calibrated reconstruction method, which does not need any explicit formulas of the kernel function for characterizing nonlinear relationships between acquired and unacquired k-space data. It is non-iterative without requiring a large amount of computational costs. Since the limited autocalibration signals (ACS) are acquired to perform KerNL calibration and the calibration suffers from the overfitting problem, more training data can improve the kernel model accuracy. In this work, virtual conjugate coil data are incorporated into the KerNL calibration and estimation process for enhancing reconstruction performance. Experimental results show that the proposed method can further suppress noise and aliasing artifacts with fewer ACS data and higher acceleration factors. Computation efficiency is still retained to keep fast reconstruction with the random projection.

Interpretable Dimension Reduction for MRI Channel Suppression.

Channel suppression can reduce the redundant information in multiple channel receiver coils and accelerate reconstruction speed to meet real-time imaging requirements. The principal component analysis has been used for channel suppression, but it is difficult to be interpreted because all channels contribute to principal components. Furthermore, the importance of interpretability in machine learning has recently attracted increasing attention in radiology. To improve the interpretability of PCA-based channel suppression, a sparse PCA method is proposed to reduce the most coils' loadings to be zero. Channel suppression is formulated as solving a nonlinear eigenvalue problem using the inverse power method instead of the direct matrix decomposition. Experimental results of in vivo data show that the sparse PCA-based channel suppression not only improves the interpretability with sparse channels, but also improves reconstruction quality compared to the standard PCA-based reconstruction with the similar reconstruction time.

A dual-interpolator method for improving parallel MRI reconstruction.

Autocalibration signal is acquired in the k-space-based parallel MRI reconstruction for estimating interpolation coefficients and reconstructing missing unacquired data. Many ACS lines can suppress aliasing artifacts and noise by covering the low-frequency signal region. However, more ACS lines will delay the data acquisition process and therefore elongate the scan time. Furthermore, a single interpolator is often used for recovering missing k-space data, and model error may exist if the single interpolator size is not selected appropriately. In this work, based on the idea of the disagreement-based semi-supervised learning, a dual-interpolator strategy is proposed to collaboratively reconstruct missing k-space data. Two interpolators with different sizes are alternatively applied to estimate and re-estimate missing data in k-space. The disagreement between two interpolators is converged and real missing values are co-estimated from two views. The experimental results show that the proposed method outperforms GRAPPA, SPIRiT, and Nonlinear GRAPPA methods using relatively low number of ACS data, and reduces aliasing artifacts and noise in reconstructed images.

Group feature selection for enhancing information gain in MRI reconstruction.

Magnetic resonance imaging (MRI) has revolutionized radiology. As a leading medical imaging modality, MRI not only visualizes the structures inside the body but also produces functional imaging. However, due to the slow imaging speed constrained by magnetic resonance physics, the MRI cost is expensive, and patients may feel not comfortable in a scanner for a long time. Parallel MRI (pMRI) has accelerated the imaging speed through a sub-Nyquist sampling strategy and the missing data are interpolated by the multiple coil data acquired. Kernel learning has been used in pMRI reconstruction to learn the interpolation weights and reconstruct the undersampled data. However, noise and aliasing artifacts still exist in the reconstructed image and a large number of auto-calibration signal lines are needed. To further improve kernel-learning-based MRI reconstruction and accelerate the speed, this paper proposes a group feature selection strategy to improve the learning performance and enhance the reconstruction quality. An explicit kernel mapping is used for selecting a subset of features which contribute most to estimating the missing k-space data. The experimental results show that the learning behaviors can be better predicted and therefore the reconstructed image quality can be improved.

Parallel MRI Reconstruction Using Broad Learning System.

As an inverse problem, parallel magnetic resonance imaging (pMRI) reconstruction accelerates imaging speed by interpolating missing k-space data from a group of phased-array coils. Deep learning methods have been used for improving pMRI reconstruction quality in recent years. However, deep learning approaches need a large amount of training data that are acquired from different hardware configurations and anatomical areas. Data distributions may be different between training data and testing data, and a long-time training is needed. In this work, we proposed a broad learning system based parallel MRI reconstruction that exploits approximation capability of one-layer neural network through broadening network structure with expanded nodes. Experimental results show that the proposed method is able to suppress noise in compared to the conventional pMRI reconstruction.

Improving Nonlinear Interpolation of K-Space Data Using Semi-Supervised Learning and Autoregressive Model.

Parallel magnetic resonance imaging (pMRI) accelerates data acquisition by undersampling k-space through an array of receiver coils. Finding accurate relationships between acquired and missing k-space data determines the interpolation performance and reconstruction quality. Autocalibration signals (ACS) are generally used to learn the interpolation coefficients for reconstructing the missing k-space data. Based on the estimation-approximation error analysis in machine learning, increasing training data size can reduce estimation error and therefore enhance generalization ability of the interpolator, but scanning time will be longer if more ACS data are acquired. We propose to augment training data using unacquired and acquired data outside of ACS region through semi-supervised learning idea and autoregressive model. Local neighbor unacquired k-space data can be used for training tasks and reducing the generalization error. Experimental results show that the proposed method outperforms the conventional methods by suppressing noise and aliasing artifacts.

EEG-Based Emotion Recognition for Modulating Social-Aware Robot Navigation.

Companion robots play an important role to accompany humans and provide emotional support, such as reducing human social isolation and loneliness. Based on recognizing human partner's mental states, a companion robot is able to dynamically adjust its behaviors, and make human-robot interaction smoother and natural. Human emotion has been recognized by many modalities like facial expression and voice. Neurophysiological signals have shown promising results in emotion recognition, since it is an innate signal of human brain which cannot be faked. In this paper, emotional state recognition using a neurophysiology method is studied to guide and modulate companion-robot navigation to enhance its social capabilities. Electroencephalogram (EEG), a type of neurophysiological signals, is used to recognize human emotional state, and then feed into a navigation path planning algorithm for controlling a companion robot's routes. Simulation results show that mobile robot presents navigation behaviors modulated by dynamic human emotional states.

Decoding Brain Activity Features to Recognize Distorted Objects.

Brain decoding is able to make human interact with an external machine or robot for assisting patient's rehabilitation. Brain generic object recognition ability can be decoded through multiple neuroimaging modalities like functional magnetic resonance imaging (fMRI). On the other hand, external machine may wrongly recognize objects due to distorted noisy or blurring images caused by many factors, and therefore deteriorate performance of brain-machine interaction. In order to create better machine, generalization capability of human brain is transferred to classifier for enhancing classification accuracy of distorted images. Since homology existing between human and machine vision has been demonstrated, through decoding neural activity features of fMRI signals into feature units of convolutional neural network layers, an enhanced object recognition method is proposed to integrate brain activity into classifier for increasing classification accuracy. Experimental results show that the proposed method is able to enhance generalization capability of distorted object recognition.

Improved gradient-echo 3D magnetic resonance imaging using compressed sensing and Toeplitz encoding with phase-scrambled RF excitation.

PURPOSE: To develop a novel three-dimensional (3D) hybrid-encoding framework using compressed sensing (CS) and Toeplitz encoding with variable phase-scrambled radio-frequency (RF) excitation, which has the following advantages: low power deposition of RF pulses, reduction of the signal dynamic range, no additional hardware requirement, and signal-to-noise ratio (SNR) improvement. METHODS: In light of the actual imaging framework of magnetic resonance imaging (MRI) scanners, we applied specially tailored RF pulses with phase-scrambled RF excitation to implement a 3D hybrid Fourier-Toeplitz encoding method based on 3D gradient-recalled echo pulse (GRASS) sequence. This method exploits Toeplitz encoding along the phase encoding direction, while keeping Fourier encoding along the readout and slice encoding directions. Phantom experiments were conducted to optimize the amplitude of specially tailored RF pulses in the 3D GRASS sequence. In vivo experiments were conducted to validate the feasibility of the proposed method, and simulations were conducted to compare the 3D hybrid-encoding method with Fourier encoding and other non-Fourier encoding methods. RESULTS: An optimized low RF amplitude was obtained in the phantom experiments. Using the optimized specially tailored RF pulses, both the watermelon and knee experiments demonstrated that the proposed method was able to preserve more image details than the conventional 3D Fourier-encoded methods at acceleration factors of 3.1 and 2.0. Additionally, SNR was improved because of no additional gradients and 3D volume encoding, when compared with single-slice scanning without 3D encoding. Simulation results demonstrated that the proposed scheme was superior to the conventional Fourier encoding method, and obtained comparative performance with other non-Fourier encoding methods in preserving details. CONCLUSIONS: We developed a practical hybrid-encoding method for 3D MRI with specially tailored RF pulses of phase-scrambled RF excitation. The proposed method improves image SNR and detail preservation compared with the conventional Fourier encoding methods. Furthermore, our proposed method exhibits superior performance in terms of detail preservation, compared with the conventional Fourier encoding method.

Improving the Otsu method for MRA image vessel extraction via resampling and ensemble learning.

Accurate extraction of vessels plays an important role in assisting diagnosis, treatment, and surgical planning. The Otsu method has been used for extracting vessels in medical images. However, blood vessels in magnetic resonance angiography (MRA) image are considered as a sparse distribution. Pixels on vessels in MRA image are considered as an imbalanced data in classification of vessels and non-vessel tissues. To extract vessels accurately, a novel method using resampling technique and ensemble learning is proposed for solving the imbalanced classification problem. Each pixel is sampled multiple times through multiple local patches within the image. Then, vessel or non-vessel tissue is determined by the ensemble voting mechanism via a p-tile algorithm. Experimental results show that the proposed method is able to outperform the traditional Otsu method by extracting vessels in MRA images more accurately.

Improving GRAPPA reconstruction using joint nonlinear kernel mapped and phase conjugated virtual coils.

To improve the reconstruction condition and alleviate the noise amplification of GRAPPA reconstruction by aggregating the phase conjugated and nonlinear kernel mapped coils with the original physical coil. Nonlinear GRAPPA (NL-GRAPPA) is a kernel-based non-iterative approach which can reduce noise-induced error in GRAPPA reconstruction. And virtual conjugate coil (VCC) embeds the conjugate symmetric property of k-space into GRAPPA data synthesis to improve reconstruction condition. This work proposed NL-VCC-GRAPPA to jointly utilize the nonlinear mapped virtual coil and phase conjugated virtual coil to further reduce noise amplification in parallel imaging. In vivo static and dynamic 2D imaging accelerated by uniform undersampling schemes were performed to evaluate the proposed method in terms of visual image quality, root-mean-square-error (RMSE), and geometry factor (g-factor). The effects of acceleration factors, calibration data size and kernel shape on the proposed model were also separately analyzed and discussed. The proposed method illustrated improved visual image quality evidenced by reduced retrospective RMSE and prospective g-factor comparing with conventional GRAPPA and the recently proposed iterative SENSE-LORAKS reconstructions. Although a larger amount of calibration data and smaller kernel size were required to stabilize the calibration of fourfold extended kernel for the proposed method, it was non-iterative and relatively insensitive to parameter adjustment in the applications. The proposed NL-VCC-extension to conventional GRAPPA brings visible improvements for imaging scenarios accelerated by the widely available uniform undersampling schemes in a practically efficient manner without iteration.

Positive-contrast susceptibility imaging based on first-order primal-dual optimization.

PURPOSE: To achieve faster reconstruction and better imaging quality of positive-contrast MRI based on the susceptibility mapping by incorporating a primal-dual (PD) formulation. METHODS: The susceptibility-based positive contrast MR technique was applied to estimate arbitrary magnetic susceptibility distributions of the metallic devices using a kernel deconvolution algorithm with a regularized ℓ1 minimization. The regularized positive-contrast inversion problem and its PD formulation were derived. The visualization of the positive contrast and convergence behavior of the PD algorithm were compared with those of the nonlinear conjugate gradient algorithm, fast iterative soft-thresholding algorithm, and alternating direction method of multipliers. These methods were tested and validated on computer simulations and phantom experiments. RESULTS: The PD approach could provide a faster reconstruction time compared with other methods. Experimental results showed that the PD algorithm could achieve comparable or even better visualization and accuracy of the metallic interventional devices in positive-contrast imaging with different SNRs and orientations to the B0 field. CONCLUSION: A susceptibility-based positive-contrast imaging technique by PD algorithm was proposed. The PD approach has more superior performance than other algorithms in terms of reconstruction time and accuracy for imaging the metallic interventional devices.

Accelerating MR Imaging via Deep Chambolle-Pock Network.

Compressed sensing (CS) has been introduced to accelerate data acquisition in MR Imaging. However, CS-MRI methods suffer from detail loss with large acceleration and complicated parameter selection. To address the limitations of existing CS-MRI methods, a model-driven MR reconstruction is proposed that trains a deep network, named CP-net, which is derived from the Chambolle-Pock algorithm to reconstruct the in vivo MR images of human brains from highly undersampled complex k-space data acquired on different types of MR scanners. The proposed deep network can learn the proximal operator and parameters among the Chambolle-Pock algorithm. All of the experiments show that the proposed CP-net achieves more accurate MR reconstruction results, outperforming state-of-the-art methods across various quantitative metrics.

Robotic Path Planning Using A* Algorithm for Automatic Navigation in Magnetic Resonance Angiography.

Magnetic resonance navigation (MRN) is an emerging research technique in recent years. The micro/nano robots existing in vessels can be driven by magnetic gradients given by MR scanner. As a non-invasive vascular imaging technique, Magnetic resonance angiography (MRA) is able to provide a vascular network of an anatomy without injection of contrast agent. In order to automatically guide and drive micro/nano robots to target in vascular network, a navigation strategy is desired. In this paper, a novel path planning algorithm based on A* search is proposed. The MRA image is preliminarily processed to extract major vessels. Then, pixel-based A* search algorithm identifies the shortest path between start point and target without human supervision. Experimental results on both of simulation image and MRA image demonstrate that the proposed method is able to accomplish path planning automatically in MRA image. That path can guide the injected micro/nano robots to navigate in the blood vessels.

Kernel Principal Component Analysis of Coil Compression in Parallel Imaging.

A phased array with many coil elements has been widely used in parallel MRI for imaging acceleration. On the other hand, it results in increased memory usage and large computational costs for reconstructing the missing data from such a large number of channels. A number of techniques have been developed to linearly combine physical channels to produce fewer compressed virtual channels for reconstruction. A new channel compression technique via kernel principal component analysis (KPCA) is proposed. The proposed KPCA method uses a nonlinear combination of all physical channels to produce a set of compressed virtual channels. This method not only reduces the computational time but also improves the reconstruction quality of all channels when used. Taking the traditional GRAPPA algorithm as an example, it is shown that the proposed KPCA method can achieve better quality than both PCA and all channels, and at the same time the calculation time is almost the same as the existing PCA method.

A survey of GPU-based acceleration techniques in MRI reconstructions.

Image reconstruction in magnetic resonance imaging (MRI) clinical applications has become increasingly more complicated. However, diagnostic and treatment require very fast computational procedure. Modern competitive platforms of graphics processing unit (GPU) have been used to make high-performance parallel computations available, and attractive to common consumers for computing massively parallel reconstruction problems at commodity price. GPUs have also become more and more important for reconstruction computations, especially when deep learning starts to be applied into MRI reconstruction. The motivation of this survey is to review the image reconstruction schemes of GPU computing for MRI applications and provide a summary reference for researchers in MRI community.

Multimodal Image Alignment via Linear Mapping between Feature Modalities.

We propose a novel landmark matching based method for aligning multimodal images, which is accomplished uniquely by resolving a linear mapping between different feature modalities. This linear mapping results in a new measurement on similarity of images captured from different modalities. In addition, our method simultaneously solves this linear mapping and the landmark correspondences by minimizing a convex quadratic function. Our method can estimate complex image relationship between different modalities and nonlinear nonrigid spatial transformations even in the presence of heavy noise, as shown in our experiments carried out by using a variety of image modalities.

Instrument Variables for Reducing Noise in Parallel MRI Reconstruction.

Generalized autocalibrating partially parallel acquisition (GRAPPA) has been a widely used parallel MRI technique. However, noise deteriorates the reconstructed image when reduction factor increases or even at low reduction factor for some noisy datasets. Noise, initially generated from scanner, propagates noise-related errors during fitting and interpolation procedures of GRAPPA to distort the final reconstructed image quality. The basic idea we proposed to improve GRAPPA is to remove noise from a system identification perspective. In this paper, we first analyze the GRAPPA noise problem from a noisy input-output system perspective; then, a new framework based on errors-in-variables (EIV) model is developed for analyzing noise generation mechanism in GRAPPA and designing a concrete method-instrument variables (IV) GRAPPA to remove noise. The proposed EIV framework provides possibilities that noiseless GRAPPA reconstruction could be achieved by existing methods that solve EIV problem other than IV method. Experimental results show that the proposed reconstruction algorithm can better remove the noise compared to the conventional GRAPPA, as validated with both of phantom and in vivo brain data.