Probabilistic Graphical Model Approach Research Articles

No Excess Mortality up to 10 Years in Early Stages of Breast Cancer in Women Adherent to Oral Endocrine Therapy: A Probabilistic Graphical Modeling Approach.

Breast cancer (BC) is globally the most frequent cancer in women. Adherence to endocrine therapy (ET) in hormone-receptor-positive BC patients is active and voluntary for the first five years after diagnosis. This study examines the impact of adherence to ET on 10-year excess mortality (EM) in patients diagnosed with Stages I to III BC (N = 2297). Since sample size is an issue for estimating age- and stage-specific survival indicators, we developed a method, ComSynSurData, for generating a large synthetic dataset (SynD) through probabilistic graphical modeling of the original cohort. We derived population-based survival indicators using a Bayesian relative survival model fitted to the SynD. Our modeling showed that hormone-receptor-positive BC patients diagnosed beyond 49 years of age at Stage I or beyond 59 years at Stage II do not have 10-year EM if they follow the prescribed ET regimen. This result calls for developing interventions to promote adherence to ET in patients with hormone receptor-positive BC and in turn improving cancer survival. The presented methodology here demonstrates the potential use of probabilistic graphical modeling for generating reliable synthetic datasets for validating population-based survival indicators when sample size is an issue.

Probabilistic elderly person's mood analysis based on its activities of daily living using smart facilities.

The world's population is aging, and eldercare services that use smart facilities such as smart homes are widely common in societies now. With the aid of smart facilities, the present study aimed at understanding an elder's moods based on the person’s activities of daily living (ADLs). With this end in view, an explainable probabilistic graphical modeling approach, applying the Bayesian network (BN), was proposed. The proposed BN-based model was capable of defining the relationship between the elder's ADLs and moods in three different levels: Activity-based Feature (AbF), Category of Activity (CoA), and the mood state. The model also allowed us to explain the transformations among the different levels/nodes on the defined BNs. A framework featured with smart facilities, including a smart home, a smartphone, and a wristband, was utilized to assess the model. The smart home was an elderly woman's house, equipped with a set of binary-based sensors. For about five months, the ADLs' data have been recorded through daily behavioral-based information, registered by experts using a defined questionnaire. The obtained results proved that the proposed BN-based model of the current study could promisingly estimate the elder's moods and CoA states. Moreover, in contrast to the machine learning techniques that behave like a black box, the effect of each feature from the lower levels to the higher levels of information of the BNs can be traced. Implications of the findings for future diagnosis and treatment of the elderly are considered.

Sequential credit card fraud detection: A joint deep neural network and probabilistic graphical model approach

AbstractWith the wide usage of e‐banking in recent years, and by increased opportunities for fraudsters subsequently, we are witnessing a loss of billions of Euros worldwide due to credit card fraud every year. Therefore, credit card fraud detection has become a critical necessity for financial institutions. Several studies have used machine learning techniques for proposing a method to address the problem. However, most of them did not take into account the sequential nature of transactional data. In this paper, we proposed a novel credit card fraud detection model using sequence labelling based on both deep neural networks and probabilistic graphical models (PGM). Then by using two real‐world datasets, we compared our model with the baseline model and examined how considering hidden sequential dependencies among transactions and also among predicted labels can improve the results. Moreover, we introduce a novel undersampling algorithm, which helps to maintain the sequential patterns of data during the random undersampling process. Our experiments demonstrate that this algorithm achieves promising results compared to the state‐of‐the‐art methods in oversampling and undersampling.

A novel approach to triple-negative breast cancer molecular classification reveals a luminal immune-positive subgroup with good prognoses

Triple-negative breast cancer is a heterogeneous disease characterized by a lack of hormonal receptors and HER2 overexpression. It is the only breast cancer subgroup that does not benefit from targeted therapies, and its prognosis is poor. Several studies have developed specific molecular classifications for triple-negative breast cancer. However, these molecular subtypes have had little impact in the clinical setting. Gene expression data and clinical information from 494 triple-negative breast tumors were obtained from public databases. First, a probabilistic graphical model approach to associate gene expression profiles was performed. Then, sparse k-means was used to establish a new molecular classification. Results were then verified in a second database including 153 triple-negative breast tumors treated with neoadjuvant chemotherapy. Clinical and gene expression data from 494 triple-negative breast tumors were analyzed. Tumors in the dataset were divided into four subgroups (luminal-androgen receptor expressing, basal, claudin-low and claudin-high), using the cancer stem cell hypothesis as reference. These four subgroups were defined and characterized through hierarchical clustering and probabilistic graphical models and compared with previously defined classifications. In addition, two subgroups related to immune activity were defined. This immune activity showed prognostic value in the whole cohort and in the luminal subgroup. The claudin-high subgroup showed poor response to neoadjuvant chemotherapy. Through a novel analytical approach we proved that there are at least two independent sources of biological information: cellular and immune. Thus, we developed two different and overlapping triple-negative breast cancer classifications and showed that the luminal immune-positive subgroup had better prognoses than the luminal immune-negative. Finally, this work paves the way for using the defined classifications as predictive features in the neoadjuvant scenario.

A Cluster Graph Approach to Land Cover Classification Boosting

When it comes to land cover classification, the process of deriving the land classes is complex due to possible errors in algorithms, spatio-temporal heterogeneity of the Earth observation data, variation in availability and quality of reference data, or a combination of these. This article proposes a probabilistic graphical model approach, in the form of a cluster graph, to boost geospatial classifications and produce a more accurate and robust classification and uncertainty product. Cluster graphs can be characterized as a means of reasoning about geospatial data such as land cover classifications by considering the effects of spatial distribution, and inter-class dependencies in a computationally efficient manner. To assess the capabilities of our proposed cluster graph boosting approach, we apply it to the field of land cover classification. We make use of existing land cover products (GlobeLand30, CORINE Land Cover) along with data from Volunteered Geographic Information (VGI), namely OpenStreetMap (OSM), to generate a boosted land cover classification and the respective uncertainty estimates. Our approach combines qualitative and quantitative components through the application of our probabilistic graphical model and subjective expert judgments. Evaluating our approach on a test region in Garmisch-Partenkirchen, Germany, our approach was able to boost the overall land cover classification accuracy by 1.4% when compared to an independent reference land cover dataset. Our approach was shown to be robust and was able to produce a diverse, feasible and spatially consistent land cover classification in areas of incomplete and conflicting evidence. On an independent validation scene, we demonstrated that our cluster graph boosting approach was generalizable even when initialized with poor prior assumptions.

Comparison of Different Hypotheses Regarding the Spread of Alzheimer’s Disease Using Markov Random Fields and Multimodal Imaging

Alzheimer's disease (AD) is characterized by a cascade of pathological processes that can be assessed in vivo using different neuroimaging methods. Recent research suggests a systematic sequence of pathogenic events on a global biomarker level, but little is known about the associations and dependencies of distinct lesion patterns on a regional level. Markov random fields are a probabilistic graphical modeling approach that represent the interaction between individual random variables by an undirected graph. We propose the novel application of this approach to study the interregional associations and dependencies between multimodal imaging markers of AD pathology and to compare different hypotheses regarding the spread of the disease. We retrieved multimodal imaging data from 577 subjects enrolled in the Alzheimer's Disease Neuroimaging Initiative. Mean amyloid load (AV45-PET), glucose metabolism (FDG-PET), and gray matter volume (MRI) were calculated for the six principle nodes of the default mode network- a functional network of brain regions that appears to be preferentially targeted by AD. Multimodal Markov random field models were developed for three different hypotheses regarding the spread of the disease: the "intraregional evolution model", the "trans-neuronal spread" hypothesis, and the "wear-and-tear" hypothesis. The model likelihood to reflect the given data was evaluated using tenfold cross-validation with 1,000 repetitions. The most likely graph structure contained the posterior cingulate cortex as main hub region with edges to various other regions, in accordance with the "wear-and-tear" hypothesis of disease vulnerability. Probabilistic graphical models facilitate the analysis of interactions between several variables in a network model and therefore afford great potential to complement traditional multiple regression analyses in multimodal neuroimaging research.

Parallel probabilistic graphical model approach for nonparametric Bayesian inference

We propose an efficient uncertainty quantification framework that makes use of multiple probabilistic graphical models to yield a nonparametric Gaussian mixture description of the target probability distribution. The methodology is indeed generic, but this work focuses on its application to the particular class of the inference problems arising from the hidden Markov process and the associated observations in a sequence. The implementation procedure is demonstrated with the dynamical system models in both low and high dimension. In case of the low dimension, it is shown that the usual factor graph for the sequential data can be used to produce a very accurate approximate solution. However, for high dimensional systems, a new family of the factor graphs are developed in order to achieve an effective dimension reduction and to facilitate a synergetic application together with multiple graphs in addressing the Bayesian data assimilation. As a result, a new paradigm for the probabilistic filtering and smoothing emerges, and the applicability of the graphical model approach has been broadened.

Abstract P6-07-07: Triple negative breast cancer classification according to cancer stem cell hypothesis

Abstract Background: Triple negative breast cancer (TNBC) is described by lack of ER and PR expression and HER2 overexpression. This subgroup has no targeted therapies and its prognosis its worse than other breast cancer subtypes. The Cancer Stem Cell hypothesis lies in two ideas; first, that breast cancers can initiate in different cell types, which can be any epithelial stem cells or any of their progeny, and, that this original cell will give the tumor a specific molecular profile. Our work proposes a division in molecular groups of TNBC looking for this tissue of origin molecular profile through gene expression data and probabilistic graphical models analyses. Material and methods: TNBC gene expression data was obtained from GSE31519 (n=494). 2000 most variable genes were selected for subsequent analysis. A functional network was built using a probabilistic graphical model approach. Functional nodes were defined, and its function was explored by Gene Ontology using DAVID. Then, a new molecular classification was generated using the activity of functional nodes and k-means. Subgroups were characterized and compared with previous TNBC molecular classifications. Results: Probabilistic graphical models defined a functional structure comprising 27 functional nodes. We found some Luminal, some Basal and a claudin-enriched node. Based on these nodes molecular subgroups were defined, following the cancer stem cell hypothesis. Thus, four subtypes: (Luminal Androgen receptor (LAR), basal, claudin-low and claudin-high were defined matching tumor origin characteristics with those in the actual tumor sample. Tumors with low expression of claudins (CLDN-low subtype) had been differentiated in the first steps of the mammary epithelial development. Tumors with high expression of claudins (CLDN-High) had been originated at the second step of the development. Next step in the development are basal-epithelial cells, which will generate Basal subtype of TNBC. And finally, the last step in development is the differentiation to luminal cell, which is the origin of the Luminal subtype. Immune status, determined by immune functional nodes, showed prognostic value (p>0.05). Finally, we compared our classification with previous ones defined by Lehmann, Burstein and the PAM50. Table 1Cellular ClassificacionTumor sizeGradeNodal T1RestG3G1 or G2N0N1Basal761631994516835CLDN-High227315194CLDN-Low103221202811LAR115429333618Total9927628010325168 Conclusion: Functional networks can provide a relevant molecular knowledge which complements the TNBC classification. From this approach we establish a new classification taking into account the cancer stem cell hypothesis. Besides, this deep knowledge will allow a more accurate prediction of outcome and can also be used for diagnostic purposes and therapy selection. Citation Format: Zamora-Auñón P, Trilla-Fuertes L, Diaz-Almiron M, Gamez-Pozo A, Prado-Vázquez G, Zapater-Moros A, Llorente-Armijo S, Gaya Romero F, Espinosa Arranz E, Fresno-Vara JA. Triple negative breast cancer classification according to cancer stem cell hypothesis [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P6-07-07.

Multivariate exploration of non-intrusive load monitoring via spatiotemporal pattern network

Non-intrusive load monitoring (NILM) of electrical demand for the purpose of identifying load components has thus far mostly been studied using univariate data, e.g., using only whole building electricity consumption time series to identify a certain type of end-use such as lighting load. However, using additional variables in the form of multivariate time series data may provide more information in terms of extracting distinguishable features in the context of energy disaggregation. In this work, a novel probabilistic graphical modeling approach, namely the spatiotemporal pattern network (STPN) is proposed for energy disaggregation using multivariate time-series data. The STPN framework is shown to be capable of handling diverse types of multivariate time-series to improve the energy disaggregation performance. The technique outperforms the state of the art factorial hidden Markov models (FHMM) and combinatorial optimization (CO) techniques in multiple real-life test cases. Furthermore, based on two homes’ aggregate electric consumption data, a similarity metric is defined for the energy disaggregation of one home using a trained model based on the other home (i.e., out-of-sample case). The proposed similarity metric allows us to enhance scalability via learning supervised models for a few homes and deploying such models to many other similar but unmodeled homes with significantly high disaggregation accuracy.

Metabolic network segmentation: A probabilistic graphical modeling approach to identify the sites and sequential order of metabolic regulation from non-targeted metabolomics data.

In recent years, the number of large-scale metabolomics studies on various cellular processes in different organisms has increased drastically. However, it remains a major challenge to perform a systematic identification of mechanistic regulatory events that mediate the observed changes in metabolite levels, due to complex interdependencies within metabolic networks. We present the metabolic network segmentation (MNS) algorithm, a probabilistic graphical modeling approach that enables genome-scale, automated prediction of regulated metabolic reactions from differential or serial metabolomics data. The algorithm sections the metabolic network into modules of metabolites with consistent changes. Metabolic reactions that connect different modules are the most likely sites of metabolic regulation. In contrast to most state-of-the-art methods, the MNS algorithm is independent of arbitrary pathway definitions, and its probabilistic nature facilitates assessments of noisy and incomplete measurements. With serial (i.e., time-resolved) data, the MNS algorithm also indicates the sequential order of metabolic regulation. We demonstrated the power and flexibility of the MNS algorithm with three, realistic case studies with bacterial and human cells. Thus, this approach enables the identification of mechanistic regulatory events from large-scale metabolomics data, and contributes to the understanding of metabolic processes and their interplay with cellular signaling and regulation processes.

A nonparametric belief propagation method for uncertainty quantification with applications to flow in random porous media

A probabilistic graphical model approach to uncertainty quantification for flows in random porous media is introduced. Model reduction techniques are used locally in the graph to represent the random permeability. Then the conditional distribution of the multi-output responses on the low dimensional representation of the permeability field is factorized into a product of local potential functions. An expectation–maximization algorithm is used to learn the nonparametric representation of these potentials using the given input/output data. We develop a nonparametric belief propagation method for uncertainty quantification by employing the loopy belief propagation algorithm. The nonparametric nature of our model is able to capture non-Gaussian features of the response. The proposed framework can be used as a surrogate model to predict the responses for new input realizations as well as our confidence on these predictions. Numerical examples are presented to demonstrate the accuracy and efficiency of the proposed framework for solving uncertainty quantification problems in flows through porous media using stationary and non-stationary permeability fields.

A probabilistic graphical model approach to stochastic multiscale partial differential equations

We develop a probabilistic graphical model based methodology to efficiently perform uncertainty quantification in the presence of both stochastic input and multiple scales. Both the stochastic input and model responses are treated as random variables in this framework. Their relationships are modeled by graphical models which give explicit factorization of a high-dimensional joint probability distribution. The hyperparameters in the probabilistic model are learned using sequential Monte Carlo (SMC) method, which is superior to standard Markov chain Monte Carlo (MCMC) methods for multi-modal distributions. Finally, we make predictions from the probabilistic graphical model using the belief propagation algorithm. Numerical examples are presented to show the accuracy and efficiency of the predictive capability of the developed graphical model.

Resilience of all-optical network architectures under in-band crosstalk attacks: a probabilistic graphical model approach

An important question for secure all-optical networks (AONs) is how to incorporate security against attacks in the design and engineering of network architectures. In this work, we study the resilience of AON architectures under in-band crosstalk attacks. Crosstalk attack propagation depends on both optical devices at the physical layer and wavelength usage at the network layer. This motivates us to employ probabilistic graphical models to model attack propagation. At the physical layer, we use a directed probabilistic graph (Bayesian Belief Network) to model the attack propagation under static network traffic and a given source of attack. At the network layer, we use an undirected probabilistic graph to represent the probability distribution of active connections in the network. The cross-layer model is obtained by combining the physical- and the network-layer models into a factor graph representation. Graphical models provide an explicit representation of interactions between the physical- and the network layer. Furthermore, graphical models facilitate derivations of analytical results on resilience with respect to physical-layer vulnerability, physical topology, and network load. Specifically, we derive bounds on the network resilience for regular topologies. For ring, star, and mesh-torus networks with link-shortest path routing and all-to-all traffic, we show that the average network resilience loss grows linearly with respect to the network load when the network load is small, and polynomially with respect to the probability of attack propagation from node to node along the attacker's route. In addition, numerical results suggest that the sum-product algorithm based on the factor graph representation can be used for computationally efficient evaluation of network resilience for irregular/large topologies