Compressed sensing based real-time dynamic MRI reconstruction

A deep unrolled neural network for real-time MRI-guided brain intervention

Accurate navigation and targeting are critical for neurological interventions including biopsy and deep brain stimulation. Real-time image guidance further improves surgical planning and MRI is ideally suited for both pre- and intra-operative imaging. However, balancing spatial and temporal resolution is a major challenge for real-time interventional MRI (i-MRI). Here, we proposed a deep unrolled neural network, dubbed as LSFP-Net, for real-time i-MRI reconstruction. By integrating LSFP-Net and a custom-designed, MR-compatible interventional device into a 3 T MRI scanner, a real-time MRI-guided brain intervention system is proposed. The performance of the system was evaluated using phantom and cadaver studies. 2D/3D real-time i-MRI was achieved with temporal resolutions of 80/732.8 ms, latencies of 0.4/3.66 s including data communication, processing and reconstruction time, and in-plane spatial resolution of 1 × 1 mm2. The results demonstrated that the proposed method enables real-time monitoring of the remote-controlled brain intervention, and showed the potential to be readily integrated into diagnostic scanners for image-guided neurosurgery.

Non-uniform sampling of similar NMR spectra and its application to studies of the interaction between alpha-synuclein and liposomes

The accelerated acquisition of multidimensional NMR spectra using sparse non-uniform sampling (NUS) has been widely adopted in recent years. The key concept in NUS is that a major part of the data is omitted during measurement, and then reconstructed using, for example, compressed sensing (CS) methods. CS requires spectra to be compressible, that is, they should contain relatively few "significant" points. The more compressible the spectrum, the fewer experimental NUS points needed in order for it to be accurately reconstructed. In this paper we show that the CS processing of similar spectra can be enhanced by reconstructing only the differences between them. Accurate reconstruction can be obtained at lower sampling levels as the difference is sparser than the spectrum itself. In many situations this method is superior to "conventional" compressed sensing. We exemplify the concept of "difference CS" with one such case-the study of alpha-synuclein binding to liposomes and its dependence on temperature. To obtain information on temperature-dependent transitions between different states, we need to acquire several dozen spectra at various temperatures, with and without the presence of liposomes. Our detailed investigation reveals that changes in the binding modes of the alpha-synuclein ensemble are not only temperature-dependent but also show non-linear behavior in their transitions. Our proposed CS processing approach dramatically reduces the number of NUS points required and thus significantly shortens the experimental time.

Efficient Near-Field Radiofrequency Imaging of Impact Damage on CFRP Materials with Learning-Based Compressed Sensing

Carbon fiber-reinforced polymer (CFRP) is a widely-used composite material that is vulnerable to impact damage. Light impact damages destroy the inner structure but barely show obvious change on the surface. As a non-contact and high-resolution method to detect subsurface and inner defect, near-field radiofrequency imaging (NRI) suffers from high imaging times. Although some existing works use compressed sensing (CS) for a faster measurement, the corresponding CS reconstruction time remains high. This paper proposes a deep learning-based CS method for fast NRI, this plugin method decreases the measurement time by one order of magnitude without hardware modification and achieves real-time imaging during CS reconstruction. A special 0/1-Bernoulli measurement matrix is designed for sensor scanning firstly, and an interpretable neural network-based CS reconstruction method is proposed. Besides real-time reconstruction, the proposed learning-based reconstruction method can further reduce the required data thus reducing measurement time more than existing CS methods. Under the same imaging quality, experimental results in an NRI system show the proposed method is 20 times faster than traditional raster scan and existing CS reconstruction methods, and the required data is reduced by more than 90% than existing CS reconstruction methods.

Application of ultrasound molecular imaging based on compressed sensing reconstruction algorithm to phase change drug-loaded PLGA nanoparticles targeting breast cancer MCF-7 Cells

Objectives:

To study the ability of aptamer-modified nano-gold rods and liquid carbon-targeted PLGA nanoparticles to target in vitro using compressed sensing reconstruction algorithm, and observe the phenomenon of mediating ultrasound / photoacoustic imaging.

Methods:

PLGA nanoparticles were prepared by a double emulsification method, and the MUC1 aptamer was connected to the PLGA nanoparticles by the carbodiimide method to obtain an “aptamer-PLGA nanoparticle” targeted phase change contrast agent. Fluorescence microscopy was used to detect the in vitro targeting of breast cancer MCF-7 cells specifically identified by it, and three control groups were set up: the ordinary nanoparticle group, the aptamer interference group, and the HELA cell group. A photoacoustic instrument was used to observe the phenomenon of enhanced ultrasound / photoacoustic signal mediated in vitro.

Results:

Many targeted nanoparticles were clustered around MCF-7 cells and bound firmly, but no specific binding was observed in the non-targeted nanoparticles group, the aptamer interference group and the HELA cell group. After the targeted nanoparticle was excited by the photoacoustic instrument, the ultrasonic signal and the photoacoustic signal were significantly enhanced compared with before the excitation.

Conclusion:

The successfully prepared targeting nanoparticles have good targeting and specificity for breast cancer MCF-7 cells, and it has obvious effects on ultrasound / photoacoustic imaging, and has the potential to become a dual-mode ultrasound / photoacoustic targeted contrast agent. The various characteristics provide experimental basis for subsequent in vivo targeting experiments and are expected to become good target diagnostic molecular probes.

3-D compressed sensing optical coherence tomography using predictive coding

We present a compressed sensing (CS) algorithm and sampling strategy for reconstructing 3-D Optical Coherence Tomography (OCT) image volumes from as little as 10% of the original data. Reconstruction using the proposed method, Denoising Predictive Coding (DN-PC), is demonstrated for five clinically relevant tissue types including human heart, retina, uterus, breast, and bovine ligament. DN-PC reconstructs the difference between adjacent b-scans in a volume and iteratively applies Gaussian filtering to improve image sparsity. An a-line sampling strategy was developed that can be easily implemented in existing Spectral-Domain OCT systems and reduce scan time by up to 90%.

Motion Compensated Dynamic MRI Reconstruction With Local Affine Optical Flow Estimation

This paper proposes a novel framework to reconstruct dynamic magnetic resonance imaging (DMRI) with motion compensation (MC). Specifically, by combining the intensity-based optical flow constraint with the traditional compressed sensing scheme, we are able to jointly reconstruct the DMRI sequences and estimate the interframe motion vectors. Then, the DMRI reconstruction can be refined through MC with the estimated motion field. By employing the coarse-to-fine multi-scale resolution strategy, we are able to update the motion field in different spatial scales. The estimated motion vectors need to be interpolated to the finest resolution scale to compensate the DMRI reconstruction. Moreover, the proposed framework is capable of handling a wide class of prior information (regularizations) for DMRI reconstruction, such as sparsity, low rank, and total variation. The formulated optimization problem is solved by a primal-dual algorithm with linesearch due to its efficiency when dealing with non-differentiable problems. Experiments on various DMRI datasets validate the reconstruction quality improvement using the proposed scheme in comparison to several state-of-the-art algorithms.

Parallel imaging and convolutional neural network combined fast MR image reconstruction: Applications in low-latency accelerated real-time imaging

PURPOSE

To develop and evaluate a parallel imaging and convolutional neural network combined image reconstruction framework for low-latency and high-quality accelerated real-time MR imaging.

METHODS

Conventional Parallel Imaging reconstruction resolved as gradient descent steps was compacted as network layers and interleaved with convolutional layers in a general convolutional neural network. All parameters of the network were determined during the offline training process, and applied to unseen data once learned. The proposed network was first evaluated for real-time cardiac imaging at 1.5 T and real-time abdominal imaging at 0.35 T, using threefold to fivefold retrospective undersampling for cardiac imaging and threefold retrospective undersampling for abdominal imaging. Then, prospective undersampling with fourfold acceleration was performed on cardiac imaging to compare the proposed method with standard clinically available GRAPPA method and the state-of-the-art L1-ESPIRiT method.

RESULTS

Both retrospective and prospective evaluations confirmed that the proposed network was able to images with a lower noise level and reduced aliasing artifacts in comparison with the single-coil based and L1-ESPIRiT reconstructions for cardiac imaging at 1.5 T, and the GRAPPA and L1-ESPIRiT reconstructions for abdominal imaging at 0.35 T. Using the proposed method, each frame can be reconstructed in less than 100 ms, suggesting its clinical compatibility.

CONCLUSION

The proposed Parallel Imaging and convolutional neural network combined reconstruction framework is a promising technique that allows low-latency and high-quality real-time MR imaging.

Thermal Light Longitudinal Correlated Imaging with Random Orthogonal Matching Pursuit Algorithm

A thermal light correlated longitudinal imaging experiment is proposed. The quasi-thermal light beam is split into two beams, a test beam and a reference beam, respectively. The light in the test beam is scattered by two amplitude objects with a specific longitudinal distance between them, while the light of the reference beam travels uninterrupted. At the end of the test and reference beams, two charge-coupled detectors (CCDs) are used to measure the intensity of the optical field. Through intensity correlation measurement the images of the two detected objects can be achieved simultaneously, only if the distance between the objects is less than the longitudinal coherent length. The theoretical analysis shows that the longitudinal coherent length is determined by both the transverse size of the incoherent thermal light source and the length of the optical path. The quality of the correlated images of the two objects is improved greatly by making use of the orthogonal matching pursuit (OMP) and the proposed variant random orthogonal matching pursuit (Random-OMP) algorithms. The experimental results show that the Random-OMP algorithm is more effective than the OMP algorithm for increasing both the visibility and continuity of the images. The experimental scenario can mimic an optical tomography imaging system, and the two objects with longitudinal distance can be taken as the two transverse layers of a three-dimensional object. The proposed Random-OMP algorithm is effective for improving the quality of the tomography image and has potential value in optical tomography imaging technology using incoherent light sources.

Causal MRI reconstruction via Kalman prediction and compressed sensing correction

This technical note addresses the problem of causal online reconstruction of dynamic MRI, i.e. given the reconstructed frames till the previous time instant, we reconstruct the frame at the current instant. Our work follows a prediction-correction framework. Given the previous frames, the current frame is predicted based on a Kalman estimate. The difference between the estimate and the current frame is then corrected based on the k-space samples of the current frame; this reconstruction assumes that the difference is sparse. The method is compared against prior Kalman filtering based techniques and Compressed Sensing based techniques. Experimental results show that the proposed method is more accurate than these and considerably faster.

Accelerated magnetic resonance thermometry in the presence of uncertainties

A model-based information theoretic approach is presented to perform the task of magnetic resonance (MR) thermal image reconstruction from a limited number of observed samples on k-space. The key idea of the proposed approach is to optimally detect samples of k-space that are information-rich with respect to a model of the thermal data acquisition. These highly informative k-space samples can then be used to refine the mathematical model and efficiently reconstruct the image. The information theoretic reconstruction was demonstrated retrospectively in data acquired during MR-guided laser induced thermal therapy (MRgLITT) procedures. The approach demonstrates that locations with high-information content with respect to a model-based reconstruction of MR thermometry may be quantitatively identified. These information-rich k-space locations are demonstrated to be useful as a guide for k-space undersampling techniques. The effect of interactively increasing the predicted number of data points used in the subsampled model-based reconstruction was quantified using the L2-norm of the distance between the subsampled and fully sampled reconstruction. Performance of the proposed approach was also compared with uniform rectilinear subsampling and variable-density Poisson disk subsampling techniques. The proposed subsampling scheme resulted in accurate reconstructions using a small fraction of k-space points, suggesting that the reconstruction technique may be useful in improving the efficiency of thermometry data temporal resolution.

Bayesian online compressed sensing

In this paper, we explore the possibilities and limitations of recovering sparse signals in an online fashion. Employing a mean field approximation to the Bayes recursion formula yields an online signal recovery algorithm that can be performed with a computational cost that is linearly proportional to the signal length per update. Analysis of the resulting algorithm indicates that the online algorithm asymptotically saturates the optimal performance limit achieved by the offline method in the presence of Gaussian measurement noise, while differences in the allowable computational costs may result in fundamental gaps of the achievable performance in the absence of noise.

3D Reconstruction of Human Laryngeal Dynamics Based on Endoscopic High-Speed Recordings

Standard laryngoscopic imaging techniques provide only limited two-dimensional insights into the vocal fold vibrations not taking the vertical component into account. However, previous experiments have shown a significant vertical component in the vibration of the vocal folds. We present a 3D reconstruction of the entire superior vocal fold surface from 2D high-speed videoendoscopy via stereo triangulation. In a typical camera-laser set-up the structured laser light pattern is projected on the vocal folds and captured at 4000 fps. The measuring device is suitable for in vivo application since the external dimensions of the miniaturized set-up barely exceed the size of a standard rigid laryngoscope. We provide a conservative estimate on the resulting resolution based on the hardware components and point out the possibilities and limitations of the miniaturized camera-laser set-up. In addition to the 3D vocal fold surface, we extended previous approaches with a G2-continuous model of the vocal fold edge. The clinical applicability was successfully established by the reconstruction of visual data acquired from 2D in vivo high-speed recordings of a female and a male subject. We present extracted dynamic parameters like maximum amplitude and velocity in the vertical direction. The additional vertical component reveals deeper insights into the vibratory dynamics of the vocal folds by means of a non-invasive method. The successful miniaturization allows for in vivo application giving access to the most realistic model available and hence enables a comprehensive understanding of the human phonation process.

Noise reduction of diffusion tensor images by sparse representation and dictionary learning

Background

The low quality of diffusion tensor image (DTI) could affect the accuracy of oncology diagnosis.

Methods

We present a novel sparse representation based denoising method for three dimensional DTI by learning adaptive dictionary with the context redundancy between neighbor slices. In this study, the context redundancy among the adjacent slices of the diffusion weighted imaging volumes is utilized to train sparsifying dictionaries. Therefore, higher redundancy could be achieved for better description of image with lower computation complexity. The optimization problem is solved efficiently using an iterative block-coordinate relaxation method.

Results

The effectiveness of our proposed method has been assessed on both simulated and real experimental DTI datasets. Qualitative and quantitative evaluations demonstrate the performance of the proposed method on the simulated data. The experiments on real datasets with different b-values also show the effectiveness of the proposed method for noise reduction of DTI.

Conclusions

The proposed approach well removes the noise in the DTI, which has high potential to be applied for clinical oncology applications.

Three-Dimensional Optical Reconstruction of Vocal Fold Kinematics Using High-Speed Video With a Laser Projection System

Vocal fold kinematics and its interaction with aerodynamic characteristics play a primary role in acoustic sound production of the human voice. Investigating the temporal details of these kinematics using high-speed videoendoscopic imaging techniques has proven challenging in part due to the limitations of quantifying complex vocal fold vibratory behavior using only two spatial dimensions. Thus, we propose an optical method of reconstructing the superior vocal fold surface in three spatial dimensions using a high-speed video camera and laser projection system. Using stereo-triangulation principles, we extend the camera-laser projector method and present an efficient image processing workflow to generate the three-dimensional vocal fold surfaces during phonation captured at 4000 frames per second. Initial results are provided for airflow-driven vibration of an ex vivo vocal fold model in which at least 75% of visible laser points contributed to the reconstructed surface. The method captures the vertical motion of the vocal folds at a high accuracy to allow for the computation of three-dimensional mucosal wave features such as vibratory amplitude, velocity, and asymmetry.

Accelerating Dynamic Cardiac MR imaging using structured sparse representation

Compressed sensing (CS) has produced promising results on dynamic cardiac MR imaging by exploiting the sparsity in image series. In this paper, we propose a new method to improve the CS reconstruction for dynamic cardiac MRI based on the theory of structured sparse representation. The proposed method user the PCA subdictionaries for adaptive sparse representation and suppresses the sparse coding noise to obtain good reconstructions. An accelerated iterative shrinkage algorithm is used to solve the optimization problem and achieve a fast convergence rate. Experimental results demonstrate that the proposed method improves the reconstruction quality of dynamic cardiac cine MRI over the state-of-the-art CS method.

Compressed Sensing MR Image Reconstruction Exploiting TGV and Wavelet Sparsity

Compressed sensing (CS) based methods make it possible to reconstruct magnetic resonance (MR) images from undersampled measurements, which is known as CS-MRI. The reference-driven CS-MRI reconstruction schemes can further decrease the sampling ratio by exploiting the sparsity of the difference image between the target and the reference MR images in pixel domain. Unfortunately existing methods do not work well given that contrast changes are incorrectly estimated or motion compensation is inaccurate. In this paper, we propose to reconstruct MR images by utilizing the sparsity of the difference image between the target and the motion-compensated reference images in wavelet transform and gradient domains. The idea is attractive because it requires neither the estimation of the contrast changes nor multiple times motion compensations. In addition, we apply total generalized variation (TGV) regularization to eliminate the staircasing artifacts caused by conventional total variation (TV). Fast composite splitting algorithm (FCSA) is used to solve the proposed reconstruction problem in order to improve computational efficiency. Experimental results demonstrate that the proposed method can not only reduce the computational cost but also decrease sampling ratio or improve the reconstruction quality alternatively.

Bayesian Compressive Sensing as Applied to Directions-of-Arrival Estimation in Planar Arrays

The Bayesian compressive sensing (BCS) is applied to estimate the directions of arrival (DoAs) of narrow-band electromagnetic signals impinging on planar antenna arrangements. Starting from the measurement of the voltages induced at the output of the array elements, the performance of the BCS-based approach is evaluated when data are acquired at a single time instant and at consecutive time instants, respectively. Different signal configurations, planar array geometries, and noise conditions are taken into account, as well.