Neurite orientation dispersion and density imaging (NODDI) enables the assessment of intracellular, extracellular and free water signals from multi-shell diffusion MRI data. It is an insightful approach to characterize brain tissue microstructure. Single-shell reconstruction for NODDI parameters has been discouraged in previous studies caused by failure when fitting, especially for the neurite density index (NDI). Here, we investigated the possibility of creating robust NODDI parameter maps with single-shell data, using the isotropic volume fraction (fISO) as prior. Prior estimation was made independent of the NODDI model constraint using a dictionary learning approach. First, we used a stochastic sparse dictionary-based network (DictNet) in predicting fISO which is trained with data obtained from in vivo and simulated diffusion MRI data. In single-shell cases, the mean diffusivity (MD) and raw T2 signal with no diffusion weighting (S0) was incorporated in the dictionary for the fISO estimation. Then, the NODDI framework was used with the known fISO to estimate the NDI and orientation dispersion index (ODI). The fISO estimated by our model was compared with other fISO estimators in the simulation. Further, using both synthetic data simulation and human data collected on a 3T scanner, we compared the performance of our dictionary-based learning prior NODDI (DLpN) with the original NODDI for both single-shell and multi-shell data. Our results suggest that DLpN derived NDI and ODI parameters for single-shell protocols are comparable with original multi-shell NODDI, and protocol with b=2000 s/mm2 performs the best (error ~5% in white and grey matter). This may allow NODDI evaluation of studies on single-shell data by multi-shell scanning of two subjects for DictNet fISO training.
Purpose: Iterative Convolutional Neural Networks (CNNs) which resemble unrolled learned iterative schemes have shown to consistently deliver state-of-the-art results for image reconstruction problems across different imaging modalities. However, because these methodes include the forward model in the architecture, their applicability is often restricted to either relatively small reconstruction problems or to problems with operators which are computationally cheap to compute. As a consequence, they have so far not been applied to dynamic non-Cartesian multi-coil reconstruction problems. Methods: In this work, we propose a CNN-architecture for image reconstruction of accelerated 2D radial cine MRI with multiple receiver coils. The network is based on a computationally light CNN-component and a subsequent conjugate gradient (CG) method which can be jointly trained end-to-end using an efficient training strategy. We investigate the proposed training-strategy and compare our method to other well-known reconstruction techniques with learned and non-learned regularization methods. Results: Our proposed method outperforms all other methods based on non-learned regularization. Further, it performs similar or better than a CNN-based method employing a 3D U-Net and a method using adaptive dictionary learning. In addition, we empirically demonstrate that even by training the network with only iteration, it is possible to increase the length of the network at test time and further improve the results. Conclusions: End-to-end training allows to highly reduce the number of trainable parameters of and stabilize the reconstruction network. Further, because it is possible to change the length of the network at test time, the need to find a compromise between the complexity of the CNN-block and the number of iterations in each CG-block becomes irrelevant.
Dimension of the encoder output (i.e., the code layer) in an autoencoder is a key hyper-parameter for representing the input data in a proper space. This dimension must be carefully selected in order to guarantee the desired reconstruction accuracy. Although overcomplete representation can address this dimension issue, the computational complexity will increase with dimension. Inspired by non-parametric methods, here, we propose a metalearning approach to increase the number of basis vectors used in dynamic sparse coding on the fly. An actor-critic algorithm is deployed to automatically choose an appropriate dimension for feature vectors regarding the required level of accuracy. The proposed method benefits from online dictionary learning and fast iterative shrinkage-thresholding algorithm (FISTA) as the optimizer in the inference phase. It aims at choosing the minimum number of bases for the overcomplete representation regarding the reconstruction error threshold. This method allows for online controlling of both the representation dimension and the reconstruction error in a dynamic framework.
Compared with single-label image classification, multi-label image classification is more practical and challenging. Some recent studies attempted to leverage the semantic information of categories for improving multi-label image classification performance. However, these semantic-based methods only take semantic information as type of complements for visual representation without further exploitation. In this paper, we present an innovative path towards the solution of the multi-label image classification which considers it as a dictionary learning task. A novel end-to-end model named Deep Semantic Dictionary Learning (DSDL) is designed. In DSDL, an auto-encoder is applied to generate the semantic dictionary from class-level semantics and then such dictionary is utilized for representing the visual features extracted by Convolutional Neural Network (CNN) with label embeddings. The DSDL provides a simple but elegant way to exploit and reconcile the label, semantic and visual spaces simultaneously via conducting the dictionary learning among them. Moreover, inspired by iterative optimization of traditional dictionary learning, we further devise a novel training strategy named Alternately Parameters Update Strategy (APUS) for optimizing DSDL, which alternately optimizes the representation coefficients and the semantic dictionary in forward and backward propagation. Extensive experimental results on three popular benchmarks demonstrate that our method achieves promising performances in comparison with the state-of-the-arts. Our codes and models have been released at {https://github.com/ZFT-CQU/DSDL}.
Recent advancements in transfer learning have made it a promising approach for domain adaptation via transfer of learned representations. This is especially when relevant when alternate tasks have limited samples of well-defined and labeled data, which is common in the molecule data domain. This makes transfer learning an ideal approach to solve molecular learning tasks. While Adversarial reprogramming has proven to be a successful method to repurpose neural networks for alternate tasks, most works consider source and alternate tasks within the same domain. In this work, we propose a new algorithm, Representation Reprogramming via Dictionary Learning (R2DL), for adversarially reprogramming pretrained language models for molecular learning tasks, motivated by leveraging learned representations in massive state of the art language models. The adversarial program learns a linear transformation between a dense source model input space (language data) and a sparse target model input space (e.g., chemical and biological molecule data) using a k-SVD solver to approximate a sparse representation of the encoded data, via dictionary learning. R2DL achieves the baseline established by state of the art toxicity prediction models trained on domain-specific data and outperforms the baseline in a limited training-data setting, thereby establishing avenues for domain-agnostic transfer learning for tasks with molecule data.
We propose K-Deep Simplex(KDS) which, given a set of data points, learns a dictionary comprising synthetic landmarks, along with representation coefficients supported on a simplex. KDS employs a local weighted $\ell_1$ penalty that encourages each data point to represent itself as a convex combination of nearby landmarks. We solve the proposed optimization program using alternating minimization and design an efficient, interpretable autoencoder using algorithm unrolling. We theoretically analyze the proposed program by relating the weighted $\ell_1$ penalty in KDS to a weighted $\ell_0$ program. Assuming that the data are generated from a Delaunay triangulation, we prove the equivalence of the weighted $\ell_1$ and weighted $\ell_0$ programs. We further show the stability of the representation coefficients under mild geometrical assumptions. If the representation coefficients are fixed, we prove that the sub-problem of minimizing over the dictionary yields a unique solution. Further, we show that low-dimensional representations can be efficiently obtained from the covariance of the coefficient matrix. Experiments show that the algorithm is highly efficient and performs competitively on synthetic and real data sets.
With the explosively increasing demands on the network capacity, throughput and number of connected wireless devices, massive connectivity is an urgent problem for the next generation wireless communications. In this paper, we propose a grant-free access protocol for massive connectivity that utilizes a large number of antennas in a base station (BS) and is expected to be widely deployed in cellular networks. The scheme consists of a sparse structure in sparse code multiple access (SCMA) and receiver processing based on dictionary learning (DL). A large number of devices can transmit data without any scheduling process. Unlike existing schemes, whose signal schedulings require a lot of overhead, the scheduling overhead required by the proposed scheme is negligible, which is attractive for resource utilization and transmission power efficiency. The numerical results show that the proposed scheme has promising performance in massive connectivity scenario of cellular networks.
Millimeter-wave (mmWave) massive multiple-input-multiple-output (mMIMO) is reported as a key enabler in the fifth-generation communication and beyond. It is customary to use a lens antenna array to transform a mmWave mMIMO channel into a beamspace where the channel exhibits sparsity. This beamspace transformation is equivalent to performing a Fourier transformation of the channel. Still, a Fourier transformation is not necessarily the optimal one, due to many reasons. Accordingly, this paper proposes using a learned sparsifying dictionary as the transformation operator leading to another beamspace. Since the dictionary is obtained by training over actual channel measurements, this transformation is shown to yield two immediate advantages. First is enhancing channel sparsity, thereby leading to more efficient pilot reduction. Second is improving the channel representation quality, and thus reducing the underlying power leakage phenomenon. Consequently, this allows for both improved channel estimation and facilitated beam selection in mmWave mMIMO. Besides, a learned dictionary is also used as the precoding operator for the same reasons. Extensive simulations under various operating scenarios and environments validate the added benefits of using learned dictionaries in improving the channel estimation quality and the beam selectivity, thereby improving the spectral efficiency.
The dynamic characteristics of functional network connectivity have been widely acknowledged and studied. Both shared and unique information has been shown to be present in the connectomes. However, very little has been known about whether and how this common pattern can predict the individual variability of the brain, i.e. "brain fingerprinting", which attempts to reliably identify a particular individual from a pool of subjects. In this paper, we propose to enhance the individual uniqueness based on an autoencoder network. More specifically, we rely on the hypothesis that the common neural activities shared across individuals may lessen individual discrimination. By reducing contributions from shared activities, inter-subject variability can be enhanced. Results show that that refined connectomes utilizing an autoencoder with sparse dictionary learning can successfully distinguish one individual from the remaining participants with reasonably high accuracy (up to 99:5% for the rest-rest pair). Furthermore, high-level cognitive behavior (e.g., fluid intelligence, executive function, and language comprehension) can also be better predicted using refined functional connectivity profiles. As expected, the high-order association cortices contributed more to both individual discrimination and behavior prediction. The proposed approach provides a promising way to enhance and leverage the individualized characteristics of brain networks.
In this work we propose a technique to remove sparse impulse noise from hyperspectral images. Our algorithm accounts for the spatial redundancy and spectral correlation of such images. The proposed method is based on the recently introduced Blind Compressed Sensing (BCS) framework, i.e. it empirically learns the spatial and spectral sparsifying dictionaries while denoising the images. The BCS framework differs from existing CS techniques - which assume the sparsifying dictionaries to be data independent, and from prior dictionary learning studies which learn the dictionary in an offline training phase. Our proposed formulation have shown over 5 dB improvement in PSNR over other techniques.
Identifying seizure activities in non-stationary electroencephalography (EEG) is a challenging task, since it is time-consuming, burdensome, and dependent on expensive human resources and subject to error and bias. A computerized seizure identification scheme can eradicate the above problems, assist clinicians and benefit epilepsy research. So far, several attempts were made to develop automatic systems to help neurophysiologists accurately identify epileptic seizures. In this research, a fully automated system is presented to automatically detect the various states of the epileptic seizure. The proposed method is based on sparse representation-based classification (SRC) theory and the proposed dictionary learning using electroencephalogram (EEG) signals. Furthermore, the proposed method does not require additional preprocessing and extraction of features which is common in the existing methods. The proposed method reached the sensitivity, specificity and accuracy of 100% in 8 out of 9 scenarios. It is also robust to the measurement noise of level as much as 0 dB. Compared to state-of-the-art algorithms and other common methods, the proposed method outperformed them in terms of sensitivity, specificity and accuracy. Moreover, it includes the most comprehensive scenarios for epileptic seizure detection, including different combinations of 2 to 5 class scenarios. The proposed automatic identification of epileptic seizures method can reduce the burden on medical professionals in analyzing large data through visual inspection as well as in deprived societies suffering from a shortage of functional magnetic resonance imaging (fMRI) equipment and specialized physician.
In this paper, we introduce a novel algorithm that can dramatically reduce the number of antenna elements needed to accurately predict the direction of arrival (DOA) for multiple input multiple output (MIMO) radar. The new proposed algorithm predicts the received signal of a large antenna setup using reduced number of antenna by using coupled dictionary learning. Hence, this enables the MIMO radar to resolve more paths, which could not be resolved by the fewer antennas. Specifically, we overcome the problem of inaccurate DOA estimation due to a small virtual array setup. For example, we can use dictionary learning to predict 100 virtual array elements using only 25. To evaluate our algorithm, we used multiple signal classification (MUSIC) as a DOA estimation technique to estimate the DOA for non coherent multiple targets. The results show that using the predicted received signal, the proposed algorithm could resolve all the targets in the scene, which could not been resolved using only the received signal from the reduced antenna setup.
The increased awareness regarding the impact of energy consumption on the environment has led to an increased focus on reducing energy consumption. Feedback on the appliance level energy consumption can help in reducing the energy demands of the consumers. Energy disaggregation techniques are used to obtain the appliance level energy consumption from the aggregated energy consumption of a house. These techniques extract the energy consumption of an individual appliance as features and hence face the challenge of distinguishing two similar energy consuming devices. To address this challenge we develop methods that leverage the fact that some devices tend to operate concurrently at specific operation modes. The aggregated energy consumption patterns of a subgroup of devices allow us to identify the concurrent operating modes of devices in the subgroup. Thus, we design hierarchical methods to replace the task of overall energy disaggregation among the devices with a recursive disaggregation task involving device subgroups. Experiments on two real-world datasets show that our methods lead to improved performance as compared to baseline. One of our approaches, Greedy based Device Decomposition Method (GDDM) achieved up to 23.8%, 10% and 59.3% improvement in terms of micro-averaged f score, macro-averaged f score and Normalized Disaggregation Error (NDE), respectively.
We propose a novel sparse dictionary learning method for planar shapes in the sense of Kendall, namely configurations of landmarks in the plane considered up to similitudes. Our shape dictionary method provides a good trade-off between algorithmic simplicity and faithfulness with respect to the nonlinear geometric structure of Kendall's shape space. Remarkably, it boils down to a classical dictionary learning formulation modified using complex weights. Existing dictionary learning methods extended to nonlinear spaces either map the manifold to a reproducing kernel Hilbert space or to a tangent space. The first approach is unnecessarily heavy in the case of Kendall's shape space and causes the geometrical understanding of shapes to be lost, while the second one induces distortions and theoretical complexity. Our approach does not suffer from these drawbacks. Instead of embedding the shape space into a linear space, we rely on the hyperplane of centered configurations, including pre-shapes from which shapes are defined as rotation orbits. In this linear space, the dictionary atoms are scaled and rotated using complex weights before summation. Furthermore, our formulation is more general than Kendall's original one: it applies to discretely-defined configurations of landmarks as well as continuously-defined interpolating curves. We implemented our algorithm by adapting the method of optimal directions combined to a Cholesky-optimized order recursive matching pursuit. An interesting feature of our shape dictionary is that it produces visually realistic atoms, while guaranteeing reconstruction accuracy. Its efficiency can mostly be attributed to a clear formulation of the framework with complex numbers. We illustrate the strong potential of our approach for the characterization of datasets of shapes up to similitudes and the analysis of patterns in deforming 2D shapes.
In many applications signals reside on the vertices of weighted graphs. Thus, there is the need to learn low dimensional representations for graph signals that will allow for data analysis and interpretation. Existing unsupervised dimensionality reduction methods for graph signals have focused on dictionary learning. In these works the graph is taken into consideration by imposing a structure or a parametrization on the dictionary and the signals are represented as linear combinations of the atoms in the dictionary. However, the assumption that graph signals can be represented using linear combinations of atoms is not always appropriate. In this paper we propose a novel representation framework based on non-linear and geometry-aware combinations of graph signals by leveraging the mathematical theory of Optimal Transport. We represent graph signals as Wasserstein barycenters and demonstrate through our experiments the potential of our proposed framework for low-dimensional graph signal representation.
There has been a growing interest in the use of data-driven regularizers to solve inverse problems associated with computational imaging systems. The convolutional sparse representation model has recently gained attention, driven by the development of fast algorithms for solving the dictionary learning and sparse coding problems for sufficiently large images and data sets. Nevertheless, this model has seen very limited application to tomographic reconstruction problems. In this paper, we present a model-based tomographic reconstruction algorithm using a learnt convolutional dictionary as a regularizer. The key contribution is the use of a data-dependent weighting scheme for the l1 regularization to construct an effective denoising method that is integrated into the inversion using the Plug-and-Play reconstruction framework. Using simulated data sets we demonstrate that our approach can improve performance over traditional regularizers based on a Markov random field model and a patch-based sparse representation model for sparse and limited-view tomographic data sets.
This paper proposes a novel algorithm for image phase retrieval, i.e., for recovering complex-valued images from the amplitudes of noisy linear combinations (often the Fourier transform) of the sought complex images. The algorithm is developed using the alternating projection framework and is aimed to obtain high performance for heavily noisy (Poissonian or Gaussian) observations. The estimation of the target images is reformulated as a sparse regression, often termed sparse coding, in the complex domain. This is accomplished by learning a complex domain dictionary from the data it represents via matrix factorization with sparsity constraints on the code (i.e., the regression coefficients). Our algorithm, termed dictionary learning phase retrieval (DLPR), jointly learns the referred to dictionary and reconstructs the unknown target image. The effectiveness of DLPR is illustrated through experiments conducted on complex images, simulated and real, where it shows noticeable advantages over the state-of-the-art competitors.
Sparse coding and dictionary learning are popular techniques for linear inverse problems such as denoising or inpainting. However in many cases, the measurement process is nonlinear, for example for clipped, quantized or 1-bit measurements. These problems have often been addressed by solving constrained sparse coding problems, which can be difficult to solve, and assuming that the sparsifying dictionary is known and fixed. Here we propose a simple and unified framework to deal with nonlinear measurements. We propose a cost function that minimizes the distance to a convex feasibility set, which models our knowledge about the nonlinear measurement. This provides an unconstrained, convex, and differentiable cost function that is simple to optimize, and generalizes the linear least squares cost commonly used in sparse coding. We then propose proximal based sparse coding and dictionary learning algorithms, that are able to learn directly from nonlinearly corrupted signals. We show how the proposed framework and algorithms can be applied to clipped, quantized and 1-bit data.
Modern high-resolution microscopes, such as the scanning tunneling microscope, are commonly used to study specimens that have dense and aperiodic spatial structure. Extracting meaningful information from images obtained from such microscopes remains a formidable challenge. Fourier analysis is commonly used to analyze the underlying structure of fundamental motifs present in an image. However, the Fourier transform fundamentally suffers from severe phase noise when applied to aperiodic images. Here, we report the development of a new algorithm based on nonconvex optimization, applicable to any microscopy modality, that directly uncovers the fundamental motifs present in a real-space image. Apart from being quantitatively superior to traditional Fourier analysis, we show that this novel algorithm also uncovers phase sensitive information about the underlying motif structure. We demonstrate its usefulness by studying scanning tunneling microscopy images of a Co-doped iron arsenide superconductor and prove that the application of the algorithm allows for the complete recovery of quasiparticle interference in this material. Our phase sensitive quasiparticle interference imaging results indicate that the pairing symmetry in optimally doped NaFeAs is consistent with a sign-changing s+- order parameter.
Objective: Hydrocephalus is a medical condition in which there is an abnormal accumulation of cerebrospinal fluid (CSF) in the brain. Segmentation of brain imagery into brain tissue and CSF (before and after surgery, i.e. pre-op vs. postop) plays a crucial role in evaluating surgical treatment. Segmentation of pre-op images is often a relatively straightforward problem and has been well researched. However, segmenting post-operative (post-op) computational tomographic (CT)-scans becomes more challenging due to distorted anatomy and subdural hematoma collections pressing on the brain. Most intensity and feature based segmentation methods fail to separate subdurals from brain and CSF as subdural geometry varies greatly across different patients and their intensity varies with time. We combat this problem by a learning approach that treats segmentation as supervised classification at the pixel level, i.e. a training set of CT scans with labeled pixel identities is employed. Methods: Our contributions include: 1.) a dictionary learning framework that learns class (segment) specific dictionaries that can efficiently represent test samples from the same class while poorly represent corresponding samples from other classes, 2.) quantification of associated computation and memory footprint, and 3.) a customized training and test procedure for segmenting post-op hydrocephalic CT images. Results: Experiments performed on infant CT brain images acquired from the CURE Children's Hospital of Uganda reveal the success of our method against the state-of-the-art alternatives. We also demonstrate that the proposed algorithm is computationally less burdensome and exhibits a graceful degradation against number of training samples, enhancing its deployment potential.