Gaussian mixture density modeling, decomposition, and applications

Biosensors, Artificial Intelligence Biosensors, False Results and Novel Future Perspectives

Medical biosensors have set the basis of medical diagnostics, and Artificial Intelligence (AI) has boosted diagnostics to a great extent. However, false results are evident in every method, so it is crucial to identify the reasons behind a possible false result in order to control its occurrence. This is the first critical state-of-the-art review article to discuss all the commonly used biosensor types and the reasons that can give rise to potential false results. Furthermore, AI is discussed in parallel with biosensors and their misdiagnoses, and again some reasons for possible false results are discussed. Finally, an expert opinion with further future perspectives is presented based on general expert insights, in order for some false diagnostic results of biosensors and AI biosensors to be surpassed.

All-atom molecular dynamics simulations of polymer and polyelectrolyte brushes

Densely grafted polymer and polyelectrolyte (PE) brushes, owing to their significant abilities to functionalize surfaces for a plethora of applications in sensing, diagnostics, current rectification, surface wettability modification, drug delivery, and oil recovery, have attracted significant attention over the past several decades. Unfortunately, most of the attention has primarily focused on understanding the properties of the grafted polymer and the PE chains with little attention devoted to studying the behavior of the brush-supported ions (counterions needed to screen the PE chains) and water molecules. Over the past few years, our group has been at the forefront of addressing this gap: we have employed all-atom molecular dynamics (MD) simulations for studying a wide variety of polymer and PE brush systems with specific attention to unraveling the properties and behavior of the brush-supported water molecules and ions. Our findings have revealed some of the most fascinating properties of such brush-supported ions and water molecules, including the most remarkable control of nanofluidic transport afforded by the specific ion and water responses induced by the PE brushes grafted on the inner walls of the nanochannel. This feature article aims to summarize some of our key contributions associated with such atomistic simulations of polymer and PE brushes and brush-supported water molecules and counterions.

Metal artefact reduction in digital tomosynthesis using component decomposition

We propose a technique for metal artefact reduction in digital tomosynthesis reconstruction. Although the problem was addressed earlier in the literature, we suggest another approach, which is, in our opinion, simpler, and easier to implement. It is a two-stage algorithm. At the first stage, attenuation images are segmented by decomposing their intensity distributions into gaussian-like components. Statistical information contained in each component is used for pixel classification. Components corresponding to metallic objects are identified, and a pixel threshold value separating regions occupied by metal objects from the rest of the image is found. Based on this value, at the second stage, a smooth mapping of image intensity is applied. This makes dense regions transparent, resulting in the artefact reduction in reconstruction. The methodology is demonstrated by several examples.

A dynamic block reward approach to improve the performance of blockchain systems

In Ethereum, miners are responsible for expanding the blockchain ledger by appending new blocks of transactions in exchange for incentives. Within the current Ethereum incentive mechanism, miners can still receive a significant amount of reward when creating non-full or even empty blocks, despite their negative impact on the system performance. We provide an extensive data-driven analysis of the impact of non-full blocks on the system performance, with the help of the BlockSim simulation tool. We collect the data for 500,000 Ethereum blocks and fit the appropriate probability distributions to the data to provide input suitable for the simulator. We show that the performance of Ethereum can be improved by over 50% if all blocks were filled with transactions. We propose an adjustment to the current Ethereum incentive model to assure the received incentive is always proportional to the block utilization level. Using our proposed approach, the incentive for non-full blocks is significantly reduced, making this behavior less attractive for miners. This implies that miners would be enforced to fill their blocks with transactions, and thus the performance is pushed to its optimal level. We show that our approach can work in practice without any crucial security issues.

Masked Autoencoder for Distribution Estimation on Small Structured Data Sets

Autoregressive models are among the most successful neural network methods for estimating a distribution from a set of samples. However, these models, such as other neural methods, need large data sets to provide good estimations. We believe that knowing structural information about the data can improve their performance on small data sets. Masked autoencoder for distribution estimation (MADE) is a well-structured density estimator, which alters a simple autoencoder by setting a set of masks on its connections to satisfy the autoregressive condition. Nevertheless, this model does not benefit from extra information that we might know about the structure of the data. This information can especially be advantageous in case of training on small data sets. In this article, we propose two autoencoders for estimating the density of a small set of observations, where the data have a known Markov random field (MRF) structure. These methods modify the masking process of MADE, according to conditional dependencies inferred from the MRF structure, to reduce either the model complexity or the problem complexity. We compare the proposed methods with some related binary, discrete, and continuous density estimators on MNIST, binarized MNIST, OCR-letters, and two synthetic data sets.

TOF-PET Image Reconstruction With Multiple Timing Kernels Applied on Cherenkov Radiation in BGO

Today Time-of-Flight (TOF), in PET scanners, assumes a single, well-defined timing resolution for all events. However, recent BGO-Cherenkov detectors, combining prompt Cherenkov emission and the typical BGO scintillation, can sort events into multiple timing kernels, best described by the Gaussian mixture models. The number of Cherenkov photons detected per event impacts directly the detector time resolution and signal rise time, which can later be used to improve the coincidence timing resolution. This work presents a simulation toolkit which applies multiple timing spreads on the coincident events and an image reconstruction that incorporates this information. A full cylindrical BGO-Cherenkov PET model was compared, in terms of contrast recovery and contrast-to-noise ratio, against an LYSO model with a time resolution of 213 ps. Two reconstruction approaches for the mixture kernels were tested: 1) mixture Gaussian and 2) decomposed simple Gaussian kernels. The decomposed model used the exact mixture component applied during the simulation. Images reconstructed using mixture kernels provided similar mean value and less noise than the decomposed. However, typically, more iterations were needed. Similarly, the LYSO model, with a single TOF kernel, converged faster than the BGO-Cherenkov with multiple kernels. The results indicate that the model complexity slows down convergence. However, due to the higher sensitivity, the contrast-to-noise ratio was 26.4% better for the BGO model.

Semiparametric Clustering: A Robust Alternative to Parametric Clustering

Clustering aims at naturally grouping the data according to the underlying data distribution. The data distribution is often estimated using a parametric or nonparametric model, e.g., Gaussian mixture or kernel density estimation. Compared with nonparametric models, parametric models are statistically stable, i.e., a small perturbation of data points leads to a small change in the estimated density. However, parametric models are highly sensitive to outliers because the data distribution is far away from the parametric assumptions in the presence of outliers. Given a parametric clustering algorithm, this paper shows how to turn this algorithm into a robust one. The idea is to modify the original parametric density into a semiparametric one. The high-density data that form the core of each cluster are modeled with the original parametric density. The low-density data are often far away from the cluster cores and may have an arbitrary shape, thus are modeled using a nonparametric density. A combination of parametric and nonparametric clustering algorithms is used to group the data modeled as a semiparametric density. From the robust statistical point of view, the proposed method has good robustness properties. We test the proposed algorithm on several synthetic and 70 UCI data sets. The results indicate that the semiparametric method could significantly improve the clustering performance.

CATS: A Tool for Clustering the Ensemble of Intrinsically Disordered Peptides on a Flat Energy Landscape

We introduce the combinatorial averaged transient structure (CATS) clustering method as a means to cluster protein structure ensembles based on the distributions of protein backbone descriptor coordinates. In our study, we use phi and psi dihedral angle coordinates of the protein backbone as descriptors due to their translational and rotational invariance. The CATS method was developed to produce unique structure ensembles that are typically difficult to obtain from flat energy landscapes using a one-dimensional separation value (e.g., RMSD cutoff). Through the use of higher-dimensional descriptor coordinates, we remedy structure resolution shortcomings of standard clustering algorithms due to large RMSD fluctuations between structures. We compare the performance of CATS to an RMSD-based clustering method GROMOS, which may not be the best choice for IDP clustering since separation quality heavily relies on cutoff values instead of energy landscape minima. We demonstrate the performance of CATS and GROMOS by analyzing the all-atom molecular dynamics trajectories of the Tau/R2(273-284) fragment in solution with TMAO and urea osmolytes from prior studies. Our study reveals that the CATS method produces more unique clusters than the GROMOS method as a result of higher-dimensional distributions of the descriptor coordinates. The cluster centers produced by CATS correspond to local minima in the multidimensional potential mean force, which generates a structure ensemble that adequately samples the energy landscape.

DEWS (DEep White matter hyperintensity Segmentation framework): A fully automated pipeline for detecting small deep white matter hyperintensities in migraineurs

Migraineurs show an increased load of white matter hyperintensities (WMHs) and more rapid deep WMH progression. Previous methods for WMH segmentation have limited efficacy to detect small deep WMHs. We developed a new fully automated detection pipeline, DEWS (DEep White matter hyperintensity Segmentation framework), for small and superficially-located deep WMHs. A total of 148 non-elderly subjects with migraine were included in this study. The pipeline consists of three components: 1) white matter (WM) extraction, 2) WMH detection, and 3) false positive reduction. In WM extraction, we adjusted the WM mask to re-assign misclassified WMHs back to WM using many sequential low-level image processing steps. In WMH detection, the potential WMH clusters were detected using an intensity based threshold and region growing approach. For false positive reduction, the detected WMH clusters were classified into final WMHs and non-WMHs using the random forest (RF) classifier. Size, texture, and multi-scale deep features were used to train the RF classifier. DEWS successfully detected small deep WMHs with a high positive predictive value (PPV) of 0.98 and true positive rate (TPR) of 0.70 in the training and test sets. Similar performance of PPV (0.96) and TPR (0.68) was attained in the validation set. DEWS showed a superior performance in comparison with other methods. Our proposed pipeline is freely available online to help the research community in quantifying deep WMHs in non-elderly adults.

Functional connectivity based parcellation of early visual cortices

Human brain can be divided into multiple brain regions based on anatomical and functional properties. Recent studies showed that resting-state connectivity can be utilized for parcellating brain regions and identifying their distinctive roles. In this study, we aimed to parcellate the primary and secondary visual cortices (V1 and V2) into several subregions based on functional connectivity and to examine the functional characteristics of each subregion. We used resting-state data from a research database and also acquired resting-state data with retinotopy results from a local site. The long-range connectivity profile and three different algorithms (i.e., K-means, Gaussian mixture model distribution, and Ward's clustering algorithms) were adopted for the parcellation. We compared the parcellation results within V1 and V2 with the eccentric map in retinotopy. We found that the boundaries between subregions within V1 and V2 were located in the parafovea, indicating that the anterior and posterior subregions within V1 and V2 corresponded to peripheral and central visual field representations, respectively. Next, we computed correlations between each subregion within V1 and V2 and intermediate and high-order regions in ventral and dorsal visual pathways. We found that the anterior subregions of V1 and V2 were strongly associated with regions in the dorsal stream (V3A and inferior parietal gyrus), whereas the posterior subregions of V1 and V2 were highly related to regions in the ventral stream (V4v and inferior temporal gyrus). Our findings suggest that the anterior and posterior subregions of V1 and V2, parcellated based on functional connectivity, may have distinct functional properties.

Cluster Prototypes and Fuzzy Memberships Jointly Leveraged Cross-Domain Maximum Entropy Clustering

The classical maximum entropy clustering (MEC) algorithm usually cannot achieve satisfactory results in the situations where the data is insufficient, incomplete, or distorted. To address this problem, inspired by transfer learning, the specific cluster prototypes and fuzzy memberships jointly leveraged (CPM-JL) framework for cross-domain MEC (CDMEC) is firstly devised in this paper, and then the corresponding algorithm referred to as CPM-JL-CDMEC and the dedicated validity index named fuzzy memberships-based cross-domain difference measurement (FM-CDDM) are concurrently proposed. In general, the contributions of this paper are fourfold: 1) benefiting from the delicate CPM-JL framework, CPM-JL-CDMEC features high-clustering effectiveness and robustness even in some complex data situations; 2) the reliability of FM-CDDM has been demonstrated to be close to well-established external criteria, e.g., normalized mutual information and rand index, and it does not require additional label information. Hence, using FM-CDDM as a dedicated validity index significantly enhances the applicability of CPM-JL-CDMEC under realistic scenarios; 3) the performance of CPM-JL-CDMEC is generally better than, at least equal to, that of MEC because CPM-JL-CDMEC can degenerate into the standard MEC algorithm after adopting the proper parameters, and which avoids the issue of negative transfer; and 4) in order to maximize privacy protection, CPM-JL-CDMEC employs the known cluster prototypes and their associated fuzzy memberships rather than the raw data in the source domain as prior knowledge. The experimental studies thoroughly evaluated and demonstrated these advantages on both synthetic and real-life transfer datasets.

Bayesian model selection for pathological neuroimaging data applied to white matter lesion segmentation

In neuroimaging studies, pathologies can present themselves as abnormal intensity patterns. Thus, solutions for detecting abnormal intensities are currently under investigation. As each patient is unique, an unbiased and biologically plausible model of pathological data would have to be able to adapt to the subject's individual presentation. Such a model would provide the means for a better understanding of the underlying biological processes and improve one's ability to define pathologically meaningful imaging biomarkers. With this aim in mind, this work proposes a hierarchical fully unsupervised model selection framework for neuroimaging data which enables the distinction between different types of abnormal image patterns without pathological a priori knowledge. Its application on simulated and clinical data demonstrated the ability to detect abnormal intensity clusters, resulting in a competitive to improved behavior in white matter lesion segmentation when compared to three other freely-available automated methods.

Features analysis for identification of date and party hubs in protein interaction network of Saccharomyces Cerevisiae

BACKGROUND

It has been understood that biological networks have modular organizations which are the sources of their observed complexity. Analysis of networks and motifs has shown that two types of hubs, party hubs and date hubs, are responsible for this complexity. Party hubs are local coordinators because of their high co-expressions with their partners, whereas date hubs display low co-expressions and are assumed as global connectors. However there is no mutual agreement on these concepts in related literature with different studies reporting their results on different data sets. We investigated whether there is a relation between the biological features of Saccharomyces Cerevisiae's proteins and their roles as non-hubs, intermediately connected, party hubs, and date hubs. We propose a classifier that separates these four classes.

RESULTS

We extracted different biological characteristics including amino acid sequences, domain contents, repeated domains, functional categories, biological processes, cellular compartments, disordered regions, and position specific scoring matrix from various sources. Several classifiers are examined and the best feature-sets based on average correct classification rate and correlation coefficients of the results are selected. We show that fusion of five feature-sets including domains, Position Specific Scoring Matrix-400, cellular compartments level one, and composition pairs with two and one gaps provide the best discrimination with an average correct classification rate of 77%.

CONCLUSIONS

We study a variety of known biological feature-sets of the proteins and show that there is a relation between domains, Position Specific Scoring Matrix-400, cellular compartments level one, composition pairs with two and one gaps of Saccharomyces Cerevisiae's proteins, and their roles in the protein interaction network as non-hubs, intermediately connected, party hubs and date hubs. This study also confirms the possibility of predicting non-hubs, party hubs and date hubs based on their biological features with acceptable accuracy. If such a hypothesis is correct for other species as well, similar methods can be applied to predict the roles of proteins in those species.

Two-stage clustering via neural networks

This paper presents a two-stage approach that is effective for performing fast clustering. First, a competitive neural network (CNN) that can harmonize mean squared error and information entropy criteria is employed to exploit the substructure in the input data by identifying the local density centers. A Gravitation neural network (GNN) then takes the locations of these centers as initial weight vectors and undergoes an unsupervised update process to group the centers into clusters. Each node (called gravi-node) in the GNN is associated with a finite attraction radius and would be attracted to a nearby centroid simultaneously during the update process, creating the Gravitation-like behavior without incurring complicated computations. This update process iterates until convergence and the converged centroid corresponds to a cluster. Compared to other clustering methods, the proposed clustering scheme is free of initialization problem and does not need to pre-specify the number of clusters. The two-stage approach is computationally efficient and has great flexibility in implementation. A fully parallel hardware implementation is very possible.

Fuzzy nonparametric DTI segmentation for robust cingulum-tract extraction

This paper presents a novel segmentation-based approach for fiber-tract extraction in diffusion-tensor (DT) images. Typical tractography methods, incorporating thresholds on fractional anisotropy and fiber curvature to terminate tracking, can face serious problems arising from partial voluming and noise. For these reasons, tractography often fails to extract thin tracts with sharp changes in orientation, e.g. the cingulum. Unlike tractography--which disregards the information in the tensors that were previously tracked--the proposed method extracts the cingulum by exploiting the statistical coherence of tensors in the entire structure. Moreover, the proposed segmentation-based method allows fuzzy class memberships to optimally extract information within partial-volumed voxels. Unlike typical fuzzy-segmentation schemes employing Gaussian models that are biased towards ellipsoidal clusters, the proposed method models the manifolds underlying the classes by incorporating nonparametric data-driven statistical models. Furthermore, it exploits the nonparametric model to capture the spatial continuity and structure of the fiber bundle. The results on real DT images demonstrate that the proposed method extracts the cingulum bundle significantly more accurately as compared to tractography.

A fuzzy, nonparametric segmentation framework for DTI and MRI analysis: with applications to DTI-tract extraction

This paper presents a novel fuzzy-segmentation method for diffusion tensor (DT) and magnetic resonance (MR) images. Typical fuzzy-segmentation schemes, e.g., those based on fuzzy C means (FCM), incorporate Gaussian class models that are inherently biased towards ellipsoidal clusters characterized by a mean element and a covariance matrix. Tensors in fiber bundles, however, inherently lie on specific manifolds in Riemannian spaces. Unlike FCM-based schemes, the proposed method represents these manifolds using nonparametric data-driven statistical models. The paper describes a statistically-sound (consistent) technique for nonparametric modeling in Riemannian DT spaces. The proposed method produces an optimal fuzzy segmentation by maximizing a novel information-theoretic energy in a Markov-random-field framework. Results on synthetic and real, DT and MR images, show that the proposed method provides information about the uncertainties in the segmentation decisions, which stem from imaging artifacts including noise, partial voluming, and inhomogeneity. By enhancing the nonparametric model to capture the spatial continuity and structure of the fiber bundle, we exploit the framework to extract the cingulum fiber bundle. Typical tractography methods for tract delineation, incorporating thresholds on fractional anisotropy and fiber curvature to terminate tracking, can face serious problems arising from partial voluming and noise. For these reasons, tractography often fails to extract thin tracts with sharp changes in orientation, such as the cingulum. The results demonstrate that the proposed method extracts this structure significantly more accurately as compared to tractography.

Generalized alternating stimulation: A novel method to reduce stimulus artifact in electrically evoked compound action potentials

Stimulus artifact is one of the main limitations when considering electrically evoked compound action potential for clinical applications. Alternating stimulation (average of recordings obtained with anodic-cathodic and cathodic-anodic bipolar stimulation pulses) is an effective method to reduce stimulus artifact when evoked potentials are recorded. In this paper we extend the concept of alternating stimulation by combining anodic-cathodic and cathodic-anodic recordings with a weight in general different to 0.5. We also provide an automatic method to obtain an estimation of the optimal weights. Comparison with conventional alternating, triphasic stimulation and masker-probe paradigm shows that the generalized alternating method improves the quality of electrically evoked compound action potential responses.

Flux Tensor Constrained Geodesic Active Contours with Sensor Fusion for Persistent Object Tracking

This paper makes new contributions in motion detection, object segmentation and trajectory estimation to create a successful object tracking system. A new efficient motion detection algorithm referred to as the flux tensor is used to detect moving objects in infrared video without requiring background modeling or contour extraction. The flux tensor-based motion detector when applied to infrared video is more accurate than thresholding "hot-spots", and is insensitive to shadows as well as illumination changes in the visible channel. In real world monitoring tasks fusing scene information from multiple sensors and sources is a useful core mechanism to deal with complex scenes, lighting conditions and environmental variables. The object segmentation algorithm uses level set-based geodesic active contour evolution that incorporates the fusion of visible color and infrared edge informations in a novel manner. Touching or overlapping objects are further refined during the segmentation process using an appropriate shape-based model. Multiple object tracking using correspondence graphs is extended to handle groups of objects and occlusion events by Kalman filter-based cluster trajectory analysis and watershed segmentation. The proposed object tracking algorithm was successfully tested on several difficult outdoor multispectral videos from stationary sensors and is not confounded by shadows or illumination variations.

A Fuzzy, Nonparametric Segmentation Framework for DTI and MRI Analysis

This paper presents a novel statistical fuzzy-segmentation method for diffusion tensor (DT) images and magnetic resonance (MR) images. Typical fuzzy-segmentation schemes, e.g. those based on fuzzy-C-means (FCM), incorporate Gaussian class models which are inherently biased towards ellipsoidal clusters. Fiber bundles in DT images, however, comprise tensors that can inherently lie on more-complex manifolds. Unlike FCM-based schemes, the proposed method relies on modeling the manifolds underlying the classes by incorporating nonparametric data-driven statistical models. It produces an optimal fuzzy segmentation by maximizing a novel information-theoretic energy in a Markov-random-field framework. For DT images, the paper describes a consistent statistical technique for nonparametric modeling in Riemannian DT spaces that incorporates two very recent works. In this way, the proposed method provides uncertainties in the segmentation decisions, which stem from imaging artifacts including noise, partial voluming, and inhomogeneity. The paper shows results on synthetic and real, DT as well as MR images.

Automated diagnosis of brain tumours astrocytomas using probabilistic neural network clustering and support vector machines

A computer-aided diagnosis system was developed for assisting brain astrocytomas malignancy grading. Microscopy images from 140 astrocytic biopsies were digitized and cell nuclei were automatically segmented using a Probabilistic Neural Network pixel-based clustering algorithm. A decision tree classification scheme was constructed to discriminate low, intermediate and high-grade tumours by analyzing nuclear features extracted from segmented nuclei with a Support Vector Machine classifier. Nuclei were segmented with an average accuracy of 86.5%. Low, intermediate, and high-grade tumours were identified with 95%, 88.3%, and 91% accuracies respectively. The proposed algorithm could be used as a second opinion tool for the histopathologists.

A Weighted Sum Validity Function for Clustering With a Hybrid Niching Genetic Algorithm

Clustering is inherently a difficult problem, both with respect to the construction of adequate objective functions as well as to the optimization of the objective functions. In this paper, we suggest an objective function called the Weighted Sum Validity Function (WSVF), which is a weighted sum of the several normalized cluster validity functions. Further, we propose a Hybrid Niching Genetic Algorithm (HNGA), which can be used for the optimization of the WSVF to automatically evolve the proper number of clusters as well as appropriate partitioning of the data set. Within the HNGA, a niching method is developed to preserve both the diversity of the population with respect to the number of clusters encoded in the individuals and the diversity of the subpopulation with the same number of clusters during the search. In addition, we hybridize the niching method with the k-means algorithm. In the experiments, we show the effectiveness of both the HNGA and the WSVF. In comparison with other related genetic clustering algorithms, the HNGA can consistently and efficiently converge to the best known optimum corresponding to the given data in concurrence with the convergence result. The WSVF is found generally able to improve the confidence of clustering solutions and achieve more accurate and robust results.

Survey of clustering algorithms

Data analysis plays an indispensable role for understanding various phenomena. Cluster analysis, primitive exploration with little or no prior knowledge, consists of research developed across a wide variety of communities. The diversity, on one hand, equips us with many tools. On the other hand, the profusion of options causes confusion. We survey clustering algorithms for data sets appearing in statistics, computer science, and machine learning, and illustrate their applications in some benchmark data sets, the traveling salesman problem, and bioinformatics, a new field attracting intensive efforts. Several tightly related topics, proximity measure, and cluster validation, are also discussed.

Automated segmentation of multiple sclerosis lesions by model outlier detection

This paper presents a fully automated algorithm for segmentation of multiple sclerosis (MS) lesions from multispectral magnetic resonance (MR) images. The method performs intensity-based tissue classification using a stochastic model for normal brain images and simultaneously detects MS lesions as outliers that are not well explained by the model. It corrects for MR field inhomogeneities, estimates tissue-specific intensity models from the data itself, and incorporates contextual information in the classification using a Markov random field. The results of the automated method are compared with lesion delineations by human experts, showing a high total lesion load correlation. When the degree of spatial correspondence between segmentations is taken into account, considerable disagreement is found, both between expert segmentations, and between expert and automatic measurements.