Expectation-maximization Framework Research Articles

Estimating Quasiperiodic Disturbance With Unknown Frequency via Expectation-Maximization.

This article is concerned with a quasiperiodic disturbance estimation problem for dynamic control systems without prior knowledge on frequency. As a major challenge of our work, the quasiperiodic disturbance to be treated is always submerged by untargeted waves, leading to complicated coupling between disturbance separation and frequency identification. Existing approaches on quasiperiodic disturbance rejection have circumvented, rather than overcome, this challenge by assuming either a known frequency or a measurable disturbance signal. In this work, an expectation-maximization (EM) framework is proposed where disturbance signal separation and frequency identification are carried out in an iterative manner. In the E-step, the expected log-likelihood function is evaluated via reconstruction of the quasiperiodic signal based on the latest frequency estimate; and in the M-step, the frequency estimate is updated by maximizing the log-likelihood function obtained in the E-step. To facilitate recursive frequency estimation, an online EM algorithm is also developed based on the forward-only smoothing techniques. Furthermore, we show that the proposed method can be easily extended to deal with nonlinear system models and time-varying frequencies.

Class-Agnostic Weighted Normalization of Staining in Histopathology Images Using a Spatially Constrained Mixture Model.

The colorless biopsied tissue samples are usually stained in order to visualize different microscopic structures for diagnostic purposes. But color variations associated with the process of sample preparation, usage of raw materials, diverse staining protocols, and using different slide scanners may adversely influence both visual inspection and computer-aided image analysis. As a result, many methods are proposed for histopathology image stain normalization in recent years. In this study, we introduce a novel approach for stain normalization based on learning a mixture of multivariate skew-normal distributions for stain clustering and parameter estimation alongside a stain transformation technique. The proposed method, labeled "Class-Agnostic Weighted Normalization" (short CLAW normalization), has the ability to normalize a source image by learning the color distribution of both source and target images within an expectation-maximization framework. The novelty of this approach is its flexibility to quantify the underlying both symmetric and nonsymmetric distributions of the different stain components while it is considering the spatial information. The performance of this new stain normalization scheme is tested on several publicly available digital pathology datasets to compare it against state-of-the-art normalization algorithms in terms of ability to preserve the image structure and information. All in all, our proposed method performed superior more consistently in comparison with existing methods in terms of information preservation, visual quality enhancement, and boosting computer-aided diagnosis algorithm performance.

MRF Model-Based Estimation of Camera Parameters and Detection of Underwater Moving Objects

The detection of underwater objects in a video is a challenging problem particularly when both the camera and the objects are in motion. In this article, this problem has been conceived as an incomplete data problem and hence the problem is formulated in expectation maximization (EM) framework. In the E-step, the frame labels are the maximum a posterior (MAP) estimates, which are obtained using simulated annealing (SA) and the iterated conditional mode (ICM) algorithm. In the M-step, the camera model parameters, both intrinsic and extrinsic, are estimated. In case of parameter estimation, the features are extracted at coarse and fine scale. In order to continuously detect the object in different video frames, EM algorithm is repeated for each frame. The performance of the proposed scheme has been compared with other algorithms and the proposed algorithm is found to outperform.

Blind Kalman Filtering for Short-Term Load Forecasting

In this work we address the problem of short-term load forecasting. We propose a generalization of the linear state-space model where the evolution of the state and the observation matrices is unknown. The proposed blind Kalman filter algorithm proceeds via alternating the estimation of these unknown matrices and the inference of the state, within the framework of expectation-maximization. A mini-batch processing strategy is introduced to allow on-the-fly forecasting. The experimental results show that the proposed method outperforms the state-of-the-art techniques by a considerable margin, both on load profile estimation and peak load forecast problems.

Contrast Pattern Mining in Paired Multivariate Time Series of a Controlled Driving Behavior Experiment

The controlled experiment is an important scientific method for researchers seeking to determine the influence of the intervention, by interpreting the contrast patterns between the temporal observations from control and experimental groups (i.e., paired multivariate time series (PMTS)). Due to recent technological advances and the growing popularity of sensing technology such as in-vehicle sensors and activity trackers, time series data is experiencing explosive growth in both size and complexity. This is threatening to overwhelm the interpretation of control experiments, which conventionally rely on human analysts. Thus, it is imperative to develop automated methods that are expected to simultaneously characterize and detect the interpretable contrast patterns in PMTS generated by controlled experiments. However, a few challenges prohibit existing methods from directly addressing this problem: (1) handling the coupling of contrast identification and pattern characterization, (2) dynamically characterizing the patterns in PMTS, and (3) mining the contrast patterns in multiple PMTS with ubiquitous individual differences. Therefore, we propose a novel framework to mine interpretable contrast patterns based on the dynamic feature dependencies for PMTS through optimization. The proposed framework simultaneously characterizes the dynamic feature dependency networks for PMTS and detects the contrast patterns. Specifically, we characterize the generative process of PMTS as a probabilistic model defined by pairwise Markov random fields whose likelihoods are maximized using our group graphical lasso. The model is then generalized to handle multiple PMTS and solved by proposing a customized algorithm based on the expectation-maximization framework. Extensive experiments demonstrate the effectiveness, scalability, and interpretability of our approach.

Semi-Blind Channel Estimation for Diffusive Molecular Communication

In this letter, we consider the problem of channel estimation for diffusive molecular communication (MC) systems. The presence of memory in diffusive MC channels, along with channel noise caused by various sources, necessitate the development of accurate channel estimators to acquire the channel impulse response (CIR). Previous works proposed pilot-based estimators based on the maximum likelihood (ML) and least squares (LS) criteria. In contrast, we propose three novel semi-blind estimators, one based on the expectation maximization (EM) framework and two based on the decision-directed (DD) estimation strategy. We also obtain the corresponding semi-blind Cramer-Rao bound (CRB). Our simulation results show that all the proposed semi-blind estimators offer substantially lower mean-squared error than the existing pilot-based estimators. The EM estimator provides the highest accuracy and converges to the semi-blind CRB, while the DD estimators offer convenient low-complexity alternatives. Importantly, the proposed estimators allow for a significant reduction in the number of transmitted pilots, without compromising the estimation accuracy.

A Variational Bayesian Causal Analysis Approach for Time-Varying Systems

Data-driven causal modeling approaches find application in process data analysis and control. Process systems can be time-varying in nature; however, the existing literature for data-driven causal modeling hardly accounts for this. In this article, we present a causal modeling approach for the time-varying systems. It relies on the time-varying parameter models (TVPMs), estimated under the variational Bayesian expectation-maximization (VBEM) framework. We incorporate a hypothesis switching procedure followed by the estimation that allows us to infer the time-varying strengths of the causal influence of the inputs on the outputs of the system. We illustrate the proposed approach using the production data from steam-assisted gravity drainage (SAGD) wells. The proposed approach was found to extract consistent causal models across multiple cases of the studied system and outperform the time-invariant models and TVPMs estimated under the maximum likelihood framework.

A mixed filter algorithm for sympathetic arousal tracking from skin conductance and heart rate measurements in Pavlovian fear conditioning.

Pathological fear and anxiety disorders can have debilitating impacts on individual patients and society. The neural circuitry underlying fear learning and extinction has been known to play a crucial role in the development and maintenance of anxiety disorders. Pavlovian conditioning, where a subject learns an association between a biologically-relevant stimulus and a neutral cue, has been instrumental in guiding the development of therapies for treating anxiety disorders. To date, a number of physiological signal responses such as skin conductance, heart rate, electroencephalography and cerebral blood flow have been analyzed in Pavlovian fear conditioning experiments. However, physiological markers are often examined separately to gain insight into the neural processes underlying fear acquisition. We propose a method to track a single brain-related sympathetic arousal state from physiological signal features during fear conditioning. We develop a state-space formulation that probabilistically relates features from skin conductance and heart rate to the unobserved sympathetic arousal state. We use an expectation-maximization framework for state estimation and model parameter recovery. State estimation is performed via Bayesian filtering. We evaluate our model on simulated and experimental data acquired in a trace fear conditioning experiment. Results on simulated data show the ability of our proposed method to estimate an unobserved arousal state and recover model parameters. Results on experimental data are consistent with skin conductance measurements and provide good fits to heartbeats modeled as a binary point process. The ability to track arousal from skin conductance and heart rate within a state-space model is an important precursor to the development of wearable monitors that could aid in patient care. Anxiety and trauma-related disorders are often accompanied by a heightened sympathetic tone and the methods described herein could find clinical applications in remote monitoring for therapeutic purposes.

Sparse Bayesian Perspective for Radar Coincidence Imaging with Model Errors

Sparsity-driven methods are commonly applied to reconstruct targets in radar coincidence imaging (RCI), where the reference matrix needs to be computed precisely and the prior knowledge of the accurate imaging model is essential. Unfortunately, the existence of model errors in practical RCI applications is common, which defocuses the reconstructed image considerably. Accordingly, this paper aims to formulate a unified framework for sparsity-driven RCI with model errors based on the sparse Bayesian approach. Firstly, a parametric joint sparse reconstruction model is built to describe the RCI when perturbed by model errors. The structured sparse Bayesian prior is then assigned to this model, after which the structured sparse Bayesian autofocus (SSBA) algorithm is proposed in the variational Bayesian expectation maximization (VBEM) framework; this solution jointly realizes sparse imaging and model error calibration. Simulation results demonstrate that the proposed algorithm can both calibrate the model errors and obtain a well-focused target image with high reconstruction accuracy.

Coarse Alignment of Topic and Sentiment: A Unified Model for Cross-Lingual Sentiment Classification.

Cross-lingual sentiment classification (CLSC) aims to leverage rich-labeled resources in the source language to improve prediction models of a resource-scarce domain in the target language. Existing feature representation learning-based approaches try to minimize the difference of latent features between different domains by exact alignment, which is achieved by either one-to-one topic alignment or matrix projection. Exact alignment, however, restricts the representation flexibility and further degrades the model performances on CLSC tasks if the distribution difference between two language domains is large. On the other hand, most previous studies proposed document-level models or ignored sentiment polarities of topics that might lead to insufficient learning of latent features. To solve the abovementioned problems, we propose a coarse alignment mechanism to enhance the model's representation by a group-to-group topic alignment into an aspect-level fine-grained model. First, we propose an unsupervised aspect, opinion, and sentiment unification model (AOS), which trimodels aspects, opinions, and sentiments of reviews from different domains and helps capture more accurate latent feature representation by a coarse alignment mechanism. To further boost AOS, we propose ps-AOS, a partial supervised AOS model, in which labeled source language data help minimize the difference of feature representations between two language domains with the help of logistics regression. Finally, an expectation-maximization framework with Gibbs sampling is then proposed to optimize our model. Extensive experiments on various multilingual product review data sets show that ps-AOS significantly outperforms various kinds of state-of-the-art baselines.

Semi-Blind Joint Timing-Offset and Channel Estimation for Amplify-and-Forward Two-Way Relaying

In this paper, we consider the problem of joint timing-offset and channel estimation for amplify-and-forward (AF) two-way relay networks (TWRNs). This problem is solved for generic pulse-shaping filters, taking into account the filter truncation in practical communication and considering both pilot-based and semi-blind estimation strategies. Beginning with pilot-based estimation, we propose a novel Maximum-likelihood joint timing-offset and channel estimator, as well as an alternative estimator based on the special properties of Zadoff-Chu sequences. The first algorithm offers high accuracy, almost overlapping with the Cramer-Rao bound (CRB), while the second offers very low computational complexity. We then develop a semi-blind estimator based on the expectation maximization (EM) framework, exploiting the underlying Hidden Markov Model to apply Baum-Welch forward-backward recursion. The semi-blind CRB is also obtained as an indicator of the best achievable performance. Using simulations, we show that the semi-blind algorithm yields superior accuracy to pilot-based estimation, as well as improved symbol-error-rates and performs very close to the semi-blind CRB. Additionally, a low-complexity approximate EM algorithm is proposed for the case of rectangular pulses. Finally, we consider the possibility of errors in integer-offset estimation and propose pilot-based and semi-blind generalized likelihood ratio test (GLRT) schemes for correcting such errors.

Compressive Holography From Poisson Noise Plagued Holograms Using Expectation-Maximization

Lensless in-line holography is a simple, portable, and cost-effective method of imaging especially for the biomedical microscopy applications. We propose a multiplicative gradient descent optimization based method to obtain multi-depth imaging from a single hologram acquired in this imaging system. We further extend the method to achieve phase imaging from a single hologram. Negative-log-likelihood functional with the assumption of poisson noise has been used as the cost function to be minimized. The ill-posed nature of the problem is handled by the sparse regularization and the upper-bound constraint. The gradient descent optimization requires calculation of the partial derivative of the cost function with respect to a given estimate of the object. A method of obtaining this quantity for holography in both the cases of real object and complex object has been shown. The reconstruction method has been validated using extensive simulation and experimental studies. The comparison with the previously established iterative shrinkage/thresholding algorithm based compressive holography shows that the proposed method has the following advantages: significantly faster convergence rate, better reconstructed image quality and the ability to perform phase imaging.

A Marked Point Process Filtering Approach for Tracking Sympathetic Arousal From Skin Conductance

Human emotion represents a complex neural process within the brain. The ability to automatically recognize emotions from physiological signals has the potential to impact humanity in multiple ways through applications in human-machine interaction, remote health monitoring, smart living environments and entertainment. We present a marked point process-based Bayesian filtering approach to track sympathetic arousal from skin conductance features. The rate at which individual skin conductance responses (SCRs) occur and their respective amplitudes encode important information regarding an individual's psychological arousal level. We develop a state-space model relating a latent neuropsychological arousal state to the rate at which neural impulses underlying SCRs occur and the impulse amplitudes. We simultaneously estimate both arousal and the state-space model parameters within an expectation-maximization framework. We evaluate our method on both simulated and experimental data. Results on simulated data indicate the method's ability to accurately estimate an unobserved state from marked point process observations. The experimental data include studies involving mental stress artificially-induced in laboratory environments, real-world driver stress and Pavlovian fear conditioning. Results on experimental data outperform previous Bayesian filtering approaches in terms of a lower sensitivity to small impulses and avoiding overfitting. The filter is thus able to estimate emotional arousal from skin conductance features and is a promising approach for everyday emotion recognition applications.

Detailed Surface Geometry and Albedo Recovery from RGB-D Video under Natural Illumination.

This article presents a novel approach for depth map enhancement from an RGB-D video sequence. The basic idea is to exploit the photometric information in the color sequence to resolve the inherent ambiguity of shape from shading problem. Instead of making any assumption about surface albedo or controlled object motion and lighting, we use the lighting variations introduced by casual object movement. We are effectively calculating photometric stereo from a moving object under natural illuminations. One of the key technical challenges is to establish correspondences over the entire image set. We, therefore, develop a lighting insensitive robust pixel matching technique that out-performs optical flow method in presence of lighting variations. An adaptive reference frame selection procedure is introduced to get more robust to imperfect lambertian reflections. In addition, we present an expectation-maximization framework to recover the surface normal and albedo simultaneously, without any regularization term. We have validated our method on both synthetic and real datasets to show its superior performance on both surface details recovery and intrinsic decomposition.

COMP-03. QUANTITATIVE IMAGE FEATURE ANALYSIS IN DIFFUSE GLIOMA – A VALUABLE MR IMAGING BIOMARKER FOR PREOPERATIVE IDH MUTATION CLASSIFICATION

Abstract INTRODUCTION Pre-operative differentiation of IDH mutant gliomas from similar appearing pathologies on imaging prior to definitive surgical diagnosis may aid treatment navigation, maximize the surgical approach, and provide diagnostic support for inoperable tumors. Quantitative image feature analysis offers a potential non-invasive method to identify diagnostic, prognostic, and predictive imaging biomarkers. We investigated the use of radiomic MR imaging features to classify tumors based on IDH mutation status. METHOD Pre-operative T1-weighted (T1W), T2-weighted (T2W), T1-contrast enhanced (T1CE), and fluid attenuated inversion recovery (FLAIR) MR brain images, along with patient IDH mutation status (mutant/wildtype) were obtained for 128 glioma patients from The Cancer Genome Atlas (TCGA). Enhancing tumor was delineated by GLISTRboost. GlistrBoost is a hybrid-discriminative model that segments tumors based on an expectation-maximization framework with a classification scheme and uses a probabilistic Bayesian strategy for segmentation refinement. MR studies for 78 glioma patients from six institutions were used for training and 50 glioma patients from a different institution were used for validation. Pre-processing included registration, resampling, and normalization. Cancer Imaging Phenomics Toolkit (CaPTK) extracted 938 radiomic image features per sequence for the enhancing tumor contour. Relevance of each individual feature was determined by the least absolute shrinkage and selection operator (LASSO). The ability of relevant radiomic image features to identify mutation status of IDH was assessed by logistic regression. RESULTS LASSO identified one highly informative radiomic imaging feature, the minimum of the mean absolute histogram deviation on T1 MR images, which was able to predict IDH mutation status with an accuracy of 0.74, precision of 1.0, and recall of 0.32. CONCLUSION Non-invasive prediction of IDH mutation status from pre-surgical MR images offers potential diagnostic, therapeutic, and prognostic benefits for glioma patients. Quantitative image feature analysis is a feasible method for identifying potential radiomic imaging features.

One-Bit Recursive Least-Squares Algorithm With Application to Distributed Target Localization

Adaptation and learning over low-cost wireless networks, meanwhile keeping an acceptable performance, are well motivated. This paper focuses on online parameter estimation over binary networks, which consist of noisy low-resolution sensors, each only giving coarsely one-bit quantized output observations and transmitting them to a fusion center. We develop a class of recursive least-squares (RLS) algorithms based on an expectation–maximization framework, which realizes adaptive parameter estimation from one-bit observations of the noisy output stream. The developed algorithms are, respectively, derived with and without prior knowledge of the noise variances, and their performances are theoretically and experimentally evaluated. Moreover, it is shown that, although the information contained in the one-bit observations is very limited, the proposed algorithms are comparable to the classical RLS algorithm using the original (nonquantized) observations. In addition, as a practical application, the proposed algorithm combined with array signal processing techniques is applied to bearing-only target localization over wireless sensor array networks, and its effectiveness is verified through simulation experiments.

Quantitative assessment of myelination patterns in preterm neonates using T2-weighted MRI

Myelination is considered to be an important developmental process during human brain maturation and closely correlated with gestational age. Quantitative assessment of the myelination status requires dedicated imaging, but the conventional T2-weighted scans routinely acquired during clinical imaging of neonates carry signatures that are thought to be associated with myelination. In this work, we develop a quatitative marker of progressing myelination for assessment preterm neonatal brain maturation based on novel automatic segmentation method for myelin-like signals on T2-weighted magnetic resonance images. Firstly we define a segmentation protocol for myelin-like signals. We then develop an expectation-maximization framework to obtain the automatic segmentations of myelin-like signals with explicit class for partial volume voxels whose locations are configured in relation to the composing pure tissues via second-order Markov random fields. The proposed segmentation achieves high Dice overlaps of 0.83 with manual annotations. The automatic segmentations are then used to track volumes of myelinated tissues in the regions of the central brain structures and brainstem. Finally, we construct a spatio-temporal growth models for myelin-like signals, which allows us to predict gestational age at scan in preterm infants with root mean squared error 1.41 weeks.

FD-MIMO via Pilot-Data Superposition: Tensor-Based DOA Estimation and System Performance

Increased pilot overhead is one of the major issues for 5G full-dimension MIMO (FD-MIMO) systems. In this paper, we introduce the FD-MIMO system using pilot-data superposition to reduce the uplink pilot overhead and investigate the impact of superimposed pilots on the overall network performance. The pilot-data superposition can relatively increase the resources allocated to the uplink data transmission. However, in time division duplex (TDD) systems, it also negatively impacts the downlink throughput since the relatively reduced pilot power can affect the estimation of the channel state information (CSI) which is used for the downlink precoding. To improve the CSI estimation, the intrinsic tensor feature of the FD-MIMO channel is exploited in our method. The introduced CSI estimation algorithm is designed through the expectation-maximization (EM) framework via tensor as the processing data structure. Furthermore, Cramer-Rao lower bound (CRLB) is utilized as a metric in our evaluation. The overall system achievable rate which is a weighted sum of the uplink and downlink throughput is investigated to reveal the fundamental trade-off between the uplink and downlink transmission in the TDD-based FD-MIMO system. In our simulation, the results demonstrate the superior performance of the introduced strategy as opposed to the conventional orthogonal pilot-data approach. Meanwhile, the pilot power allocation, as well as different transmitting and receiving strategies are investigated to offer various trade-off points between the uplink and downlink transmission.

Skin Conductance as a Viable Alternative for Closing the Deep Brain Stimulation Loop in Neuropsychiatric Disorders.

Markers from local field potentials, neurochemicals, skin conductance, and hormone concentrations have been proposed as a means of closing the loop in Deep Brain Stimulation (DBS) therapy for treating neuropsychiatric and movement disorders. Developing a closed-loop DBS controller based on peripheral signals would require: (i) the recovery of a biomarker from the source neural stimuli underlying the peripheral signal variations; (ii) the estimation of an unobserved brain or central nervous system related state variable from the biomarker. The state variable is application-specific. It is emotion-related in the case of depression or post-traumatic stress disorder, and movement-related for Parkinson's or essential tremor. We present a method for closing the DBS loop in neuropsychiatric disorders based on the estimation of sympathetic arousal from skin conductance measurements. We deconvolve skin conductance via an optimization formulation utilizing sparse recovery and obtain neural impulses from sympathetic nerve fibers stimulating the sweat glands. We perform this deconvolution via a two-step coordinate descent procedure that recovers the sparse neural stimuli and estimates physiological system parameters simultaneously. We next relate an unobserved sympathetic arousal state to the probability that these neural impulses occur and use Bayesian filtering within an Expectation-Maximization framework for estimation. We evaluate our method on a publicly available data-set examining the effect of different types of stress on peripheral signal changes including body temperature, skin conductance and heart rate. A high degree of arousal is estimated during cognitive tasks, as are much lower levels during relaxation. The results demonstrate the ability to decode psychological arousal from neural activity underlying skin conductance signal variations. The complete pipeline from recovering neural stimuli to decoding an emotion-related brain state using skin conductance presents a promising methodology for the ultimate realization of a closed-loop DBS controller. Closed-loop DBS treatment would additionally help reduce unnecessary power consumption and improve therapeutic gains.

Super-Resolution Channel Estimation for Massive MIMO via Clustered Sparse Bayesian Learning

This correspondence paper provides a novel super-resolution downlink channel estimation approach for massive multiple-input multiple-output (MIMO) systems, by jointly learning the parametric dictionary and recovering the sparse channel components. Specifically, we exploit a Markov spike and slab prior to characterize the clustered sparse channel structure resulting from small local scatterers in the angular domain. The proposed algorithm is developed within a variational expectation maximization framework and integrated with the generalized approximate message passing technique to calculate the intractable posterior distribution. Simulation results illustrate that our approach attains a significant performance improvement over existing methods.