Gaussian Mixture Model Approach Research Articles

Improvement of the Gaussian mixture models' unsupervised learning method through the inclusion of dynamical systems for various types of nonlinear data

Gaussian mixture models (GMM) with a modulating dynamical system (DS) approach is an unsupervised learning method, and it can estimate the distribution of given data or encoding trajectories in the input space. In this paper, a series of trajectories is considered for simulation, and the role of tuning parameters in the algorithm for both Gaussian function encoding and behavior of the dynamical system is obtained and compared. This algorithm divides the input space of the data into presupposed local regions and then in each local region of the data employs a dynamical system approach for tracking the major trajectories of the data. In this paper, the influence of the number of the Gaussian function in the GMM approach is investigated and simulated deeply. Furthermore, the influence of the local statistical characteristic of data such as mean or covariance of the data on the training process is discussed, and in these conditions, the effect of tuning parameters as the number of the Gaussian function is explained. Also, all details of the characteristic of DS depend on these tuning parameters, especially when data has more variance or noise, this adjustment should be checked more accurately. So, eventually, we showed in the obtained simulation results that the behavior and location of attractor points in DS on the data distributions and accordingly stability of the DS is getting improved drastically by tuning the number of Gaussian functions accurately.

A Data-Driven Model with Hysteresis Compensation for I2RIS Robot.

Retinal microsurgery is a high-precision surgery performed on a delicate tissue requiring the skill of highly trained surgeons. Given the restricted range of instrument motion in the confined intraocular space, snake-like robots may prove to be a promising technology to provide surgeons with greater flexibility, dexterity, and positioning accuracy during retinal procedures such as retinal vein cannulation and epiretinal membrane peeling. Kinematics modeling of these robots is an essential step toward accurate position control. Unlike conventional manipulators, modeling these robots does not follow a straightforward method due to their complex mechanical structure and actuation mechanisms. The hysteresis problem can especially impact the positioning accuracy significantly in wire-driven snake-like robots. In this paper, we propose a data-driven kinematics model using a probabilistic Gaussian mixture model (GMM) and Gaussian mixture regression (GMR) approach with a hysteresis compensation algorithm. Experimental results on the two-degree-of-freedom (DOF) integrated robotic intraocular snake (I2RIS) show that the proposed model with the hysteresis compensation can predict the snake tip bending angle for pitch and yaw with 0.45° and 0.39° root mean square error (RMSE), respectively. This results in overall 60% and 70% improvements of accuracy for yaw and pitch over the same model without the hysteresis compensation.

Machine learning of ECG waveforms and cardiac magnetic resonance for response and survival after cardiac resynchronization therapy

Cardiac resynchronization therapy (CRT) can lead to marked symptom reduction and improved survival in selected patients with heart failure with reduced ejection fraction (HFrEF); however, many candidates for CRT based on clinical guidelines do not have a favorable response. A better way to identify patients expected to benefit from CRT that applies machine learning to accessible and cost-effective diagnostic tools such as the 12-lead electrocardiogram (ECG) could have a major impact on clinical care in HFrEF by helping providers personalize treatment strategies and avoid delays in initiation of other potentially beneficial treatments. This study addresses this need by demonstrating that a novel approach to ECG waveform analysis using functional principal component decomposition (FPCD) performs better than measures that require manual ECG analysis with the human eye and also at least as well as a previously validated but more expensive approach based on cardiac magnetic resonance (CMR). Analyses are based on five-fold cross validation of areas under the curve (AUCs) for CRT response and survival time after the CRT implant using Cox proportional hazards regression with stratification of groups using a Gaussian mixture model approach. Furthermore, FPCD and CMR predictors are shown to be independent, which demonstrates that the FPCD electrical findings and the CMR mechanical findings together provide a synergistic model for response and survival after CRT. In summary, this study provides a highly effective approach to prognostication after CRT in HFrEF using an accessible and inexpensive diagnostic test with a major expected impact on personalization of therapies.

Optimal 2D audio features estimation for a lightweight application in mosquitoes species: Ecoacoustics detection and classification purposes

Mosquitoes are the vector of diseases that kill more than one million people per year worldwide. Surveillance systems are essential for understanding their complex ecology and behaviour. This is fundamental for predicting disease risk caused by mosquitoes and formulating effective control strategies against mosquito-borne diseases such as malaria, dengue, and Zika. Mosquito populations vary heterogeneously in urban and rural landscapes, fluctuating with seasonal and climatic trends and human activity. Several approaches provide environmental data for mosquito mapping and risk prediction. However, they rely traditionally upon labour-intensive techniques such as manual traps. This paper presents the optimal audio features for mosquito identification using ecoacoustics signals to automatically identify different mosquito species from their wingbeat sounds based on popular audio features. The audio selection method uses Density-Based Spatial Clustering of Applications with Noise (DBSCAN) and Silhouette coefficient to evaluate the clusters in the data through the optimal-combined audio features. To classify the mosquito species and distinguish them from environmental-urban noise, the method comprises the Gaussian Mixture Model (GMM) and Gibbs approach for Aedes aegypti, and Culex quinquefasciatus, using the acoustic recordings of their wingbeat signals. Finally, comparing GMM and Gibbs, the two have very similar accuracy, but the classification time is much faster for Gibbs sampling, making it a good candidate for a lightweight solution. These are essential when deploying the described models to monitor mosquito vectors in the wild with Internet of Things (IoT) technologies.

Mixed infections in genotypic drug-resistant Mycobacterium tuberculosis

Tuberculosis disease (TB), caused by Mycobacterium tuberculosis, is a major global public health problem, resulting in more than 1 million deaths each year. Drug resistance (DR), including multi-drug (MDR-TB), is making TB control difficult and accounts for 16% of new and 48% of previously treated cases. To further complicate treatment decision-making, many clinical studies have reported patients harbouring multiple distinct strains of M. tuberculosis across the main lineages (L1 to L4). The extent to which drug-resistant strains can be deconvoluted within mixed strain infection samples is understudied. Here, we analysed M. tuberculosis isolates with whole genome sequencing data (n = 50,723), which covered the main lineages (L1 9.1%, L2 27.6%, L3 11.8%, L4 48.3%), with genotypic resistance to isoniazid (HR-TB; n = 9546 (29.2%)), rifampicin (RR-TB; n = 7974 (24.4%)), and at least MDR-TB (n = 5385 (16.5%)). TB-Profiler software revealed 531 (1.0%) isolates with potential mixed sub-lineage infections, including some with DR mutations (RR-TB 21/531; HR-TB 59/531; at least MDR-TB 173/531). To assist with the deconvolution of such mixtures, we adopted and evaluated a statistical Gaussian Mixture model (GMM) approach. By simulating 240 artificial mixtures of different ratios from empirical data across L1 to L4, a GMM approach was able to accurately estimate the DR profile of each lineage, with a low error rate for the estimated mixing proportions (mean squared error 0.012) and high accuracy for the DR predictions (93.5%). Application of the GMM model to the clinical mixtures (n = 531), found that 33.3% (188/531) of samples consisted of DR and sensitive lineages, 20.2% (114/531) consisted of lineages with only DR mutations, and 40.6% (229/531) consisted of lineages with genotypic pan-susceptibility. Overall, our work demonstrates the utility of combined whole genome sequencing data and GMM statistical analysis approaches for providing insights into mono and mixed M. tuberculosis infections, thereby potentially assisting diagnosis, treatment decision-making, drug resistance and transmission mapping for infection control.

Data Preprocessing and Machine Learning Modeling for Rockburst Assessment

Rockbursts pose a significant threat to human safety and environmental stability. This paper aims to predict rockburst intensity using a machine learning model. A dataset containing 344 rockburst cases was collected, with eight inducing features as input and four rockburst grades as output. In the preprocessing stage, missing feature values were estimated using a regression imputation strategy. A novel approach, which combines feature selection (FS), t-distributed stochastic neighbor embedding (t-SNE), and Gaussian mixture model (GMM) clustering, was proposed to relabel the dataset. The effectiveness of this approach was compared with common statistical methods, and its underlying principles were analyzed. A voting ensemble strategy was used to build the machine learning model, and optimal hyperparameters were determined using the tree-structured Parzen estimator (TPE), whose efficiency and accuracy were compared with three common optimization algorithms. The best combination model was determined using performance evaluation and subsequently applied to practical rockburst prediction. Finally, feature sensitivity was studied using a relative importance analysis. The results indicate that the FS + t-SNE + GMM approach stands out as the optimum data preprocessing method, significantly improving the prediction accuracy and generalization ability of the model. TPE is the most effective optimization algorithm, characterized simultaneously by both high search capability and efficiency. Moreover, the elastic energy index Wet, the maximum circumferential stress of surrounding rock σθ, and the uniaxial compression strength of rock σc were identified as relatively important features in the rockburst prediction model.

Spatiotemporal Interaction Pattern Recognition and Risk Evolution Analysis During Lane Changes

In complex lane change (LC) scenarios, semantic interpretation and safety analysis of dynamic interaction pattern are necessary for autonomous vehicles to make appropriate decisions. This study proposes a learning framework that combines primitive-based interaction pattern recognition and risk analysis. The Hidden Markov Model with the Gaussian mixture model (GMM-HMM) approach is developed to decompose the LC scenarios into primitives. Then K-means clustering with Dynamic Time Warping (DTW) is applied to gather the primitives into 13 LC interaction patterns. Finally, this study considers time-to-collision (TTC) of two conflict types involved in the LC process. And the TTC is used to analyze the risk of interaction patterns and extract high-risk LC interaction patterns. The LC events obtained from the Highway Drone Dataset (highD) demonstrate that the identified LC interaction patterns contain interpretable semantic information. This study identifies the dynamic spatiotemporal characteristics and risk formation mechanism of the LC interaction patterns. The findings are useful to comprehensively understand the latent interaction patterns, which can then be used to design and improve the decision-making process during lane changes and enhance the safety of autonomous vehicle.

Comparison Approaches of the Fuzzy C-Means and Gaussian Mixture Model in Clustering the Welfare of the Indonesian People

We compared the previous study about clustering the welfare of the Indonesian people using the Fuzzy C-Means (FCM) approach to a recent study, the Gaussian mixture model (GMM). Both of which were soft clustering. The case analyzes by classifying 34 provincial data in Indonesia, based on eight welfare of people indicator variables in 2017, which the Central Statistics Agency had issued. We compared the FCM and the GMM approaches to determine a better level of accuracy in clustering data using the Silhouette index, the Davies-Bouldin index, and the Calinski-Harabasz index values as a validity test method. The FCM and GMM methods found that the optimal clusters were 2 and 6. When we observed the consistency of the three tests’ validity results, the GMM method was preferable to the FCM clustering method. Keywords: fuzzy, Gaussian mixture model, clustering

Weighted-indexed semi-Markov model: calibration and application to financial modeling

We address the calibration issues of the weighted-indexed semi-Markov chain (WISMC) model applied to high-frequency financial data. Specifically, we propose to automate the discretization of the price returns and the volatility index by using four different approaches, two based on statistical quantities, namely, the quantile and sigma discretization, and two derived by the application of two popular machine learning algorithms, namely the k-means and Gaussian mixture model (GMM). Moreover, by comparing the Bayesian information criterion (BIC) scores, the GMM approach allows for the selection of the number of states of returns and index. An application to Bitcoin prices at 1-min and 1-s intervals shows the validity and usefulness of the proposed discretization approaches. In particular, GMM discretization is well suited for high-frequency returns, whereas the quantile approach works better for low-frequency intervals. Finally, by comparing the results of the Monte Carlo simulation, we show that the WISMC model, applied with the proposed discretization, can reproduce the long-range serial correlation of the squared returns, which is typical of the financial markets and, in particular, the cryptocurrency market.

Probabilistic simulation of biogas production from anaerobic co-digestion using Anaerobic Digestion Model No. 1: A case study on agricultural residue

Probabilistic simulation of biogas production from anaerobic co-digestion using Anaerobic Digestion Model No. 1: A case study on agricultural residue

Feature Fusion-Based Inconsistency Evaluation for Battery Pack: Improved Gaussian Mixture Model

The large-scale grouping of the battery system leads to the inconsistency of the battery pack. Aiming at tacking this issue, an inconsistency evaluation method is deployed for the battery pack based on an improved Gaussian mixture model (GMM) and feature fusion approach. Specifically, the proposed adaptive forgetting factor recursive least squares (AFFRLS) algorithm allows the open-circuit voltage and other parameters to be jointly identified without the open circuit voltage-state of charge (OCV-SOC) test. An online capacity estimation approach with the extended Kalman particle filter (EPF) is put forward for capacity estimation. Further, an improved GMM is proposed to visualize battery pack inconsistency, using the K-means++ algorithm to initialize category centers. The standard deviation coefficient approach quantifies the inconsistency. Finally, the real-life vehicle data are performed to validate the effectiveness of the proposed method. The experimental results show that the proposed method can evaluate the battery parameters accurately. With the increase in service time, the inconsistency of the battery pack is gradually deteriorating.

A Triple High Throughput Screening for Extracellular Vesicle Inducing Agents With Immunostimulatory Activity.

Extracellular vesicles (EVs) play an important role in intercellular communication and regulation of cells, especially in the immune system where EVs can participate in antigen presentation and may have adjuvant effects. We aimed to identify small molecule compounds that can increase EV release and thereby enhance the immunogenicity of vaccines. We utilized a THP-1 reporter cell line engineered to release EV-associated tetraspanin (CD63)-Turbo-luciferase to quantitatively measure EVs released in culture supernatants as a readout of a high throughput screen (HTS) of 27,895 compounds. In parallel, the cytotoxicity of the compounds was evaluated by PrestoBlue dye assay. For screening immunostimulatory potency, we performed two additional independent HTS on the same compound library using NF-κB and interferon-stimulated response element THP-1 reporter cell lines. Hit compounds were then identified in each of the 3 HTS’s, using a “Top X″ and a Gaussian Mixture Model approach to rule out false positive compounds and to increase the sensitivity of the hit selection. Thus, 644 compounds were selected as hits which were further evaluated for induction of IL-12 in murine bone-marrow derived dendritic cells (mBMDCs) and for effects of cell viability. The resulting 130 hits were then assessed from a medicinal chemistry perspective to remove compounds with functional group liabilities. Finally, 80 compounds were evaluated as vaccine adjuvants in vivo using ovalbumin as a model antigen. We analyzed 18 compounds with adjuvant activity for their ability to induce the expression of co-stimulatory molecules on mBMDCs. The full complement of data was then used to cluster the compounds into 4 distinct biological activity profiles. These compounds were also evaluated for quantitation of EV release and spider plot overlays were generated to compare the activity profiles of compounds within each cluster. This tiered screening process identified two compounds that belong to the 4-thieno-2-thiopyrimidine scaffold with identical screening profiles supporting data reproducibility and validating the overall screening process. Correlation patterns in the adjuvanticity data suggested a role for CD63 and NF-κB pathways in potentiating antigen-specific antibody production. Thus, our three independent cell-based HTS campaigns led to identification of immunostimulatory compounds that release EVs and have adjuvant activity.

Utilizing Gaussian mixture models in all-sky searches for short-duration gravitational wave bursts

Coherent WaveBurst is a generic, multidetector gravitational wave burst search based on the excess power approach. The coherent WaveBurst algorithm currently employed in the all-sky short-duration gravitational wave burst search uses a conditional approach on selected attributes in the multidimensional event attribute space to distinguish between noisy events from that of astrophysical origin. We have been developing a supervised machine learning approach based on the Gaussian mixture modeling to model the attribute space for signals as well as noise events to enhance the probability of burst detection [Gayathri et al.Phys. Rev. D 102, 104023 (2020)]. We further extend the GMM approach to the all-sky short-duration coherent WaveBurst search as a postprocessing step on events from the first half of the third observing run (O3a). We show an improvement in sensitivity to generic gravitational wave burst signal morphologies as well as the astrophysical source such as core-collapse supernova models due to the application of our Gaussian mixture model approach to coherent WaveBurst triggers. The Gaussian mixture model method recovers the gravitational wave signals from massive compact binary coalescences identified by coherent WaveBurst targeted for binary black holes in GWTC-2, with better significance than the all-sky coherent WaveBurst search. No additional significant gravitational wave bursts are observed.

Coverage hole detection using social spider optimized Gaussian Mixture Model

Wireless sensors are backed by their battery life. The battery depletion creates a communication hole in the network. The replacement of certain energy-depleted sensor nodes necessitates their detection. The work in this article presents a novel solution for the detection of coverage hole in the WSN. An undirected graph was employed in this study to divide the WSN region into two parts: the outer and inner nodes. The Gaussian Mixture Model (GMM) clustering detects the outer/edge nodes at the coverage hole area. Node degree, closeness, betweenness, and page rank are the four graph metrics that GMM uses for clustering in this work. GMM’s clustering performance is governed by the static mean and covariance. A novel combination of social spider optimization and GMM is then leveraged to fine-tune the cluster to decrease false detection rates of holes. Distinct holes, such as round, E-shape, and rectangular, are used to conduct the analysis. Using the SSO-EM GMM approach, the comparison shows that it outperforms the current methods of coverage hole identification.

Underwater object detection based on bi-dimensional empirical mode decomposition and Gaussian Mixture Model approach

Monitoring objects and observing changes in the undersea environment are important for exploring resources and studying ecology. These conditions are challenging because the depth of the sea varies the transmission of light in the water medium intensely. Many object detection algorithms exist so far to observe the changes and for the detection of foreground objects. A new approach of underwater object detection is proposed to differentiate the foreground object from its background. It evolves with the idea of blob generation based on features extracted from underwater scenes using the Bi-dimensional Empirical Mode Decomposition (BEMD) Algorithm. The features such as weight factor, Hurst exponent, and texture strength decide the blob. These extracted features are then accustomed to the generic Gaussian Mixture Model (GMM) to provide a basis for an object monitoring system. The proposed model is primarily developed and tested for fish detection in underwater sequences. The proposed approach is compared with state-of-art object detection schemes and analysed both qualitatively and quantitatively. • An adapted empirical mode decomposition algorithm extracts the Texture strength, Weight factor, and Hurst Exponent. • The extracted features of BEMD are applied to the challenging underwater scenes to estimate the foreground objects. • The GMM-based detection, in conjunction with the extracted BEMD features, detect the object in complex environments. • The estimated objects with a temporal guided adaptive GMM provide good detection efficiency with reduced error.

Improving the ripple classification in focal pediatric epilepsy: identifying pathological high-frequency oscillations by Gaussian mixture model clustering

Objective. High-frequency oscillations (HFOs) have emerged as a promising clinical biomarker for presurgical evaluation in childhood epilepsy. HFOs are commonly classified in stereo-encephalography as ripples (80–200 Hz) and fast ripples (200–500 Hz). Ripples are less specific and not so directly associated with epileptogenic activity because of their physiological and pathological origin. The aim of this paper is to distinguish HFOs in the ripple band and to improve the evaluation of the epileptogenic zone (EZ). Approach. This study constitutes a novel modeling approach evaluated in ten patients from Sant Joan de Deu Pediatric Hospital (Barcelona, Spain), with clearly-defined seizure onset zones (SOZ) during presurgical evaluation. A subject-by-subject basis analysis is proposed: a probabilistic Gaussian mixture model (GMM) based on the combination of specific ripple features is applied for estimating physiological and pathological ripple subpopulations. Main Results. Clear pathological and physiological ripples are identified. Features differ considerably among patients showing within-subject variability, suggesting that individual models are more appropriate than a traditional whole-population approach. The difference in rates inside and outside the SOZ for pathological ripples is significantly higher than when considering all the ripples. These significant differences also appear in signal segments without epileptiform activity. Pathological ripple rates show a sharp decline from SOZ to non-SOZ contacts and a gradual decrease with distance. Significance. This novel individual GMM approach improves ripple classification and helps to refine the delineation of the EZ, as well as being appropriate to investigate the interaction of epileptogenic and propagation networks.

A generalized pH acceleration model of nano-sol products and the effects of model misspecification on shelf-life prediction

In existing pH acceleration models, which are used to assess the shelf-life of liquid-phase nano-sol products, a mixture of normal distributions is commonly employed to describe the sizes of particles mixing from two populations. The Gaussian mixture model approach falls short when used to characterize the asymmetric distributions of particle sizes in subgroups. This work considers instead a broader class of a mixture of log-F distributions, to be embedded in a pH acceleration model. This study aims at understanding the impact of the new modeling approach, in the presence of model misspecification of the particle size distribution, on the accuracy and precision for making the shelf-life predictions. This study found that model misspecification indeed significantly affects the shelf-life predictions. The proposed method shows favorable finite sample performance on a simulated data set. Both the quantitative analysis of the impact due to model misspecification and the solution proposed herein could benefit practitioners in the long run.

A Social Media Study on Mental Health Status Transitions Surrounding Psychiatric Hospitalizations.

For people diagnosed with a mental illness, psychiatric hospitalization is one step in a long journey, consisting of clinical recovery such as removal of symptoms, and social reintegration involving resuming social roles and responsibilities, overcoming stigma and self-maintenance of the condition. Both clinical recovery and social reintegration need to go hand-in-hand for the overall well-being of individuals. However, research exploring social media for mental health has considered narrower, disjoint conceptualizations of people with mental illness - either as a patient or as a support-seeker. In this paper, we combine medical records with social media data of 254 consented individuals who have experienced a psychiatric hospitalization to address this gap. Adopting a theory-driven, Gaussian Mixture modeling approach, we provide a taxonomy of six heterogeneous behavioral patterns characterizing peoples' mental health status transitions around hospitalizations. Then we present an empirically derived framework, based on feedback from clinical researchers, to understand peoples' trajectories around clinical recovery and social reintegration. Finally, to demonstrate the utility of this taxonomy and the empirical framework, we assess social media signals that are indicative of individuals' reintegration trajectories post-hospitalization. We discuss the implications of combining peoples' clinical and social experiences in mental health care and the opportunities this intersection presents to post-discharge support and technology-based interventions for mental health.

Solution strategy based on Gaussian mixture models and dispersion reduction for the capacitated centered clustering problem.

The Capacitated Centered Clustering Problem (CCCP)—a multi-facility location model—is very important within the logistics and supply chain management fields due to its impact on industrial transportation and distribution. However, solving the CCCP is a challenging task due to its computational complexity. In this work, a strategy based on Gaussian mixture models (GMMs) and dispersion reduction is presented to obtain the most likely locations of facilities for sets of client points considering their distribution patterns. Experiments performed on large CCCP instances, and considering updated best-known solutions, led to estimate the performance of the GMMs approach, termed as Dispersion Reduction GMMs, with a mean error gap smaller than 2.6%. This result is more competitive when compared to Variable Neighborhood Search, Simulated Annealing, Genetic Algorithm and CKMeans and faster to achieve when compared to the best-known solutions obtained by Tabu-Search and Clustering Search.

Brainwave Classification for Character-Writing Application using EMD-based GMM and KELM approaches

A brainwave classification, which does not involve any limb movement and stimulus for character-writing applications, benefits impaired people, in terms of practical communication, because it allows users to command a device/computer directly via electroencephalogram signals. In this paper, we propose a new framework based on Empirical Mode Decomposition (EMD) features along with the Gaussian Mixture Model (GMM) and Kernel Extreme Learning Machine (KELM)-based classifiers. For this purpose, firstly, we introduce EMD to decompose EEG signals into Intrinsic Mode Functions (IMFs), which actually are used as the input features of the brainwave classification for the character-writing application. We hypothesize that EMD along with the appropriate IMF is quite powerful for the brainwave classification, in terms of character applications, because of the wavelet-like decomposition without any down sampling process. Secondly, by getting motivated with shallow learning classifiers, we can provide promising performance for the classification of binary classes, GMM and KELM, which are applied for the learning of features along with the brainwave classification. Lastly, we propose a new method by combining GMM and KELM to fuse the merits of different classifiers. Moreover, the proposed methods are validated by using the volunteer-independent 5-fold cross-validation and accuracy as a standard measurement. The experimental results showed that EMD with the proper IMF achieved better results than the conventional discrete wavelet transform (DWT) feature. Moreover, we found that the EMD feature along with the GMM/KELM-based classifier provides the average accuracy of 77.40% and 80.10%, respectively, which could perform better than the conventional methods where we use DWT along with the artificial neural network classifier in order to get the average accuracy of 80.60%. Furthermore, we obtained the improved performance by combining GMM and KELM, i.e., average accuracy of 80.60%. These outcomes exhibit the usefulness of the EMD feature combining with GMM and KELM based classifiers for the brainwave classification in terms of the Character-Writing application, which do not require any limb movement and stimulus.