We present the first acquisition-agnostic diffusion generative model for Magnetic Resonance Imaging (MRI) in the neonatal intensive care unit (NICU) to solve a range of inverse problems for shortening scan time and improving motion robustness. In-NICU MRI scanners leverage permanent magnets at lower field-strengths (i.e., below 1.5 Tesla) for non-invasive assessment of potential brain abnormalities during the critical phase of early live development, but suffer from long scan times and motion artifacts. In this setting, training data sizes are small and intrinsically suffer from low signal-to-noise ratio (SNR). This work trains a diffusion probabilistic generative model using such a real-world training dataset of clinical neonatal MRI by applying several novel signal processing and machine learning methods to handle the low SNR and low quantity of data. The model is then used as a statistical image prior to solve various inverse problems at inference time without requiring any retraining. Experiments demonstrate the generative model’s utility for three real-world applications of neonatal MRI: accelerated reconstruction, motion correction, and super-resolution.

Magnetic Resonance Imaging (MRI) is highly susceptible to motion artifacts due to the extended acquisition times required for k-space sampling. These artifacts can compromise diagnostic utility, particularly for dynamic imaging. We propose a novel alternating minimization framework that leverages a bespoke diffusion model to jointly reconstruct and correct non-rigid motion-corrupted k-space data. The diffusion model uses a coarse-to-fine denoising strategy to capture large overall motion and reconstruct the lower frequencies of the image first, providing a better inductive bias for motion estimation than that of standard diffusion models. We demonstrate the performance of our approach on both real-world cine cardiac MRI datasets and complex simulated rigid and non-rigid deformations, even when each motion state is undersampled by a factor of 64×. Additionally, our method is agnostic to sampling patterns, anatomical variations, and MRI scanning protocols, as long as some low frequency components are sampled during each motion state.

We provide a framework for solving inverse problems with diffusion models learned from linearly corrupted data. Firstly, we extend the Ambient Diffusion framework to enable training directly from measurements corrupted in the Fourier domain. Subsequently, we train diffusion models for MRI with access only to Fourier subsampled multi-coil measurements at acceleration factors R$=2, 4, 6, 8$. Secondly, we propose $\textit{Ambient Diffusion Posterior Sampling}$ (A-DPS), a reconstruction algorithm that leverages generative models pre-trained on one type of corruption (e.g. image inpainting) to perform posterior sampling on measurements from a different forward process (e.g. image blurring). For MRI reconstruction in high acceleration regimes, we observe that A-DPS models trained on subsampled data are better suited to solving inverse problems than models trained on fully sampled data. We also test the efficacy of A-DPS on natural image datasets (CelebA, FFHQ, and AFHQ) and show that A-DPS can sometimes outperform models trained on clean data for several image restoration tasks in both speed and performance.

Abstract. This work aims to accelerate the convergence of proximal gradient methods used to solve regularized linear inverse problems. This is achieved by designing a polynomial-based preconditioner that targets the eigenvalue spectrum of the normal operator derived from the linear operator. The preconditioner does not assume any explicit structure on the linear function and thus can be deployed in diverse applications of interest. The efficacy of the preconditioner is validated on three different Magnetic Resonance Imaging applications, where it is seen to achieve faster iterative convergence (around [Formula: see text] faster, depending on the application of interest) while achieving similar reconstruction quality.

Channel estimation is a critical task in multiple-input multiple-output (MIMO) digital communications that substantially affects end-to-end system performance. In this work, we introduce a novel approach for channel estimation using deep score-based generative models. A model is trained to estimate the gradient of the logarithm of a distribution and is used to iteratively refine estimates given measurements of a signal. We introduce a framework for training score-based generative models for wireless MIMO channels and performing channel estimation based on posterior sampling at test time. We derive theoretical robustness guarantees for channel estimation with posterior sampling in single-input single-output scenarios, and experimentally verify performance in the MIMO setting. Our results in simulated channels show competitive in-distribution performance, and robust out-of-distribution performance, with gains of up to 5 dB in end-to-end coded communication performance compared to supervised deep learning methods. Simulations on the number of pilots show that high fidelity channel estimation with 25% pilot density is possible for MIMO channel sizes of up to$64 \times 256$. Complexity analysis reveals that model size can efficiently trade performance for estimation latency, and that the proposed approach is competitive with compressed sensing in terms of floating-point operation (FLOP) count.

Channel estimation is a critical task in digital communications that greatly impacts end-to-end system performance. In this work, we introduce a novel approach for multiple-input multiple-output (MIMO) channel estimation using score-based generative models. Our method uses a deep neural network that is trained to estimate the gradient of the log-prior of wireless channels at any point in high-dimensional space, and leverages this model to solve channel estimation via posterior sampling. We train a score-based model on channel realizations from the CDL-D model for two antenna spacings and show that the approach leads to competitive in- and out-of-distribution performance when compared to generative adversarial network (GAN) and compressed sensing (CS) methods. When tested on CDL-D channels, the approach leads to a gain of at least 5 dB in channel estimation error compared to GAN methods in-distribution at λ/2 antenna spacing. When tested on CDL-C channels which are never seen during training or fine-tuned on, the approach leads to end-to-end coded performance gains of up to 3 dB compared to CS methods and losses of only 0.5 dB compared to ideal channel knowledge.

Deep learning has been recently applied to physical layer processing in digital communication systems in order to improve end-to-end performance. In this work, we introduce a novel deep learning solution for soft bit quantization across wideband channels. Our method is trained end-to-end with quantization-and entropy-aware augmentations to the loss function and is used at inference in conjunction with source coding to achieve near-optimal compression gains over wideband channels. We prove and verify that a proper weight initialization scheme leads to a reduced feature variance, which allows us to use a fixed latent feature quantization scheme. When tested on channel distributions never seen during training, the proposed method achieves a compression gain of up to 10% in the high SNR regime versus previous state-of-the-art methods.

Efficient modeling of relational data arising in physical, social, and information sciences is challenging due to complicated dependencies within the data. In this work we build off of semi-implicit graph variational auto-encoders to capture higher order statistics in a low-dimensional graph latent representation. We incorporate hyperbolic geometry in the latent space through a Poincare embedding to efficiently represent graphs exhibiting hierarchical structure. To address the naive posterior latent distribution assumptions in classical variational inference, we use semi-implicit hierarchical variational Bayes to implicitly capture posteriors of given graph data, which may exhibit heavy tails, multiple modes, skewness, and highly correlated latent structures. We show that the existing semi-implicit variational inference objective provably reduces information in the observed graph. Based on this observation, we estimate and add an additional mutual information term to the semi-implicit variational inference learning objective to capture rich correlations arising between the input and latent spaces. We show that the inclusion of this regularization term in conjunction with the \poincare embedding boosts the quality of learned high-level representations and enables more flexible and faithful graphical modeling. We experimentally demonstrate that our approach outperforms existing graph variational auto-encoders both in Euclidean and in hyperbolic spaces for edge link prediction and node classification.

The CSGM framework (Bora-Jalal-Price-Dimakis'17) has shown that deepgenerative priors can be powerful tools for solving inverse problems.However, to date this framework has been empirically successful only oncertain datasets (for example, human faces and MNIST digits), and itis known to perform poorly on out-of-distribution samples. In thispaper, we present the first successful application of the CSGMframework on clinical MRI data. We train a generative prior on brainscans from the fastMRI dataset, and show that posterior sampling viaLangevin dynamics achieves high quality reconstructions. Furthermore,our experiments and theory show that posterior sampling is robust tochanges in the ground-truth distribution and measurement process.Our code and models are available at: \url{https://github.com/utcsilab/csgm-mri-langevin}.

Implementing the Iterative Soft-Thresholding Algorithm (ISTA) of compressed sensing for MRI image reconstruction is a good candidate for designing accelerators because real-time functional MRI applications require intensive computations. A straightforward mapping of the computation graph of ISTA onto an FPGA, with a wide enough datapath to saturate memory bandwidth, would require substantial resources, such that a modest size FPGA would not fit the reconstruction pipeline for an entire MRI image. This paper proposes several methods to design the kernel components of ISTA, such as matrix transpose, datapath reuse, parallelism within maps, and data buffering to overcome the problem. Our implementation with Intel OpenCL SDK and performance evaluation on Intel HARPv2 show that our methods can map the reconstruction for the entire 256 × 256 MRI image with 8 or more channels to its FPGA, while achieving good overall performance.

Linear operators used in iterative methods like conjugate gradient have typically been implemented either as ""matrix-driven"" subroutines backed by explicit sparse or dense matrices, or as ""matrix-free"" subroutines that implement specific linear operations directly (e.g. FFTs). The matrix-driven approach is generally more portable because it can target widely-available BLAS libraries, but it can be inefficient in terms of time and space complexity. In contrast, the matrix-free approach is more performant because it leverages structure in operations, but it requires each operator be re-implemented on each new platform. To increase performance and portability, we propose a hybrid approach that represents linear operators as expression trees. Leaf nodes in the tree are either matrix-free or matrix-driven operators, and interior nodes represent mathematical compositions (sums, products, transposes) or structural compositions (stacks, block diagonals, etc.) of the leaf operators. This representation enables expert-guided reordering and fusion transformations that can improve performance or reduce memory pressure. We implement our approach in a domain-specific language called Indigo. We assess Indigo on image reconstruction problems arising in four application areas: magnetic resonance imaging, ptychography, magnetic particle imaging, and fluorescent microscopy. We give performance results from vendor BLAS libraries, and we introduce specializations to Sparse BLAS routines that achieve near-Roofline performance on multi-core, many-core, and GPU systems.

Index coding, initially introduced within theoretical computer science to address a specialized class of problems, has gained significant interest within communications and networking communities in recent years. Index coding has been shown to be analogous to a large class of challenging wired network coding and wireless multi-terminal problems, the latter class being of primary interest in this paper. Here, a (relaxed) rank minimization based analytic framework is presented for wireless index coding, which represents a first step in a systematic algorithmic approach to index coding for practical use. Further, the paper demonstrates its applicability over a real-world wireless testbed. The scheme operates at the network layer, and can be understood as a (non-trivial) generalization of existing principles of random linear network coding. Experimental results demonstrate that, for a class of network topologies, the rank-minimized index coding system presents a throughput gain of 50 to 100 percent greater than random linear coding for this system.

Massively broadband® RF channel sounding is severely constrained by the sampling rates required for analog to digital conversion. Analog compressed sensing (CS) techniques based on Xampling have demonstrated the ability to lower sampling rates far below the Nyquist rate. Here, we show attributes of the multipath channel sounding problem appear to be well suited to CS approaches for reducing measurement acquisition time while simultaneously estimating time delays, multipath amplitudes, and angles of arrival. This paper presents results of the fusion of CS with modern channel sounding. We show measured propagation data from 60 GHz field trials and note the channel sparsity in time and space. We then propose an architecture for the first massively broadband CS channel sounder based on the Xampling framework (which we call the Channel Sounding Xampler) to exploit the sparsity, and we use field measurements to explore tradeoffs between analog and digital signal processing to perform channel impulse response (CIR) parameter estimation in real time. We also offer conceptual approaches for the Channel Sounding Xampler designed to trade off analog and digital components with the goal of improving CIR acquisition at sub-THz frequencies.

Wireless systems require increasingly large system bandwidths that are only available at millimeter-wave frequencies. Such spectrum bands offer the potential for multi-gigabit-per-second data rates to low-cost massively broadband® devices. To enable mobile outdoor millimeter-wave cellular-type applications, it is necessary to determine the coverage potential of base stations in real-world environments. This paper presents the results of a measurement campaign of 38 GHz outdoor urban cellular channels using directional antennas at both the mobile and the base station, and assesses outage probabilities at two separate transmitter locations on the campus of The University of Texas at Austin. Our measurements demonstrate the viability of directional antennas and site-specific planning for future mm-wave cellular, and show that cell radii of ~200 M will provide a very high probability of coverage in an urban environment. As production costs for millimeter-wave technologies continue to fall [1], we envision millimeter-wave cellular systems with dense base station deployments as a cost effective means of delivering multi-Gbps data rates to mobile cell phone and internet users.