Joint Bayesian Model Research Articles

Predicting trawl catches using environmental DNA

Abstract Quantifying the biomass, or number of individuals, diversity, and distribution of marine species is a critical aspect of understanding and managing marine ecosystems. In recent years, there has been growing interest in using environmental DNA (eDNA) for marine ecosystem management and biodiversity assessment. However, the main challenge hindering eDNA applicability has been the inability to infer absolute species abundances from multispecies analysis (eDNA metabarcoding). In this study, we demonstrate a way forward by estimating the abundance of commercially important fish species in a Norwegian fjord using a joint Bayesian statistical model of traditional trawl-catch data and molecular data derived from eDNA. Using this model, we accurately predict out-of-sample trawl catches using eDNA alone. Moreover, our model provides empirical estimates for key processes linking marine eDNA concentration to the fish population abundance estimated from trawl observations, including trawl catchability, DNA shedding, degradation, dilution, transport, recovery rate, and isolation efficiency. These processes, including amplification efficiencies correcting for Polymerase Chain Reaction (PCR) bias, are species-specific and enable the translation of eDNA metabarcoding data into abundances. These findings have broad implications for the use of eDNA in marine ecosystem management and conservation efforts.

Geospatial joint modeling of vector and parasite serology to microstratify malaria transmission

The World Health Organization identifies a strong surveillance system for malaria and its mosquito vector as an essential pillar of the malaria elimination agenda. Anopheles salivary antibodies are emerging biomarkers of exposure to mosquito bites that potentially overcome sensitivity and logistical constraints of traditional entomological surveys. Using samples collected by a village health volunteer network in 104 villages in Southeast Myanmar during routine surveillance, the present study employs a Bayesian geostatistical modeling framework, incorporating climatic and environmental variables together with Anopheles salivary antigen serology, to generate spatially continuous predictive maps of Anopheles biting exposure. Our maps quantify fine-scale spatial and temporal heterogeneity in Anopheles salivary antibody seroprevalence (ranging from 9 to 99%) that serves as a proxy of exposure to Anopheles bites and advances current static maps of only Anopheles occurrence. We also developed an innovative framework to perform surveillance of malaria transmission. By incorporating antibodies against the vector and the transmissible form of malaria (sporozoite) in a joint Bayesian geostatistical model, we predict several foci of ongoing transmission. In our study, we demonstrate that antibodies specific for Anopheles salivary and sporozoite antigens are a logistically feasible metric with which to quantify and characterize heterogeneity in exposure to vector bites and malaria transmission. These approaches could readily be scaled up into existing village health volunteer surveillance networks to identify foci of residual malaria transmission, which could be targeted with supplementary interventions to accelerate progress toward elimination.

Bayesian relative composite quantile regression with ordinal longitudinal data and some case studies

ABSTRACT In real applied fields such as clinical medicine, environmental sciences, psychology as well as economics, we often encounter the task of conducting statistical inference for longitudinal data with ordinal responses. The traditional methods of longitudinal data analysis are often inclined to model continuous responses, which are no longer suitable for such ordinal data. Logistic regression and probit regression are two considerable methods which are frequently used to model ordinal longitudinal responses. However, such modelling methods just depict the mean feature of latent outcome variable and may produce non-robust results when encountering nor-normal errors or outliers. As a proper alternative of mean regression models, composite quantile regression (CQR) method is usually employed to derive robust estimation. The target of this paper is to investigate the CQR estimation approach for ordinal latent longitudinal model. The joint Bayesian hierarchical model is established and a relative CQR estimation approach is suggested to conduct posterior inference for the considered model. Further, in longitudinal data modelling, excessive predictors may be brought into in the models which result in the decrease of the model prediction precision. Bayesian L 1 / 2 regularized prior is incorporated into ordinal longitudinal CQR model to conduct variable selection simultaneously. Finally, simulation studies and two ordinal longitudinal data analysis are hired to illustrate the considered method.

The Association between Mediated Deprivation and Ovarian Cancer Survival among African American Women.

Deprivation indices are often used to adjust for socio-economic disparities in health studies. Their role has been partially evaluated for certain population-level cancer outcomes, but examination of their role in ovarian cancer is limited. In this study, we evaluated a range of well-recognized deprivation indices in relation to cancer survival in a cohort of self-identified Black women diagnosed with ovarian cancer. This study aimed to determine if clinical or diagnostic characteristics lie on a mediating pathway between socioeconomic status (SES) and deprivation and ovarian cancer survival in a minority population that experiences worse survival from ovarian cancer. We used mediation analysis to look at the direct and indirect causal effects of deprivation indices with main mediators of the SEER stage at diagnosis and residual disease. The analysis employed Bayesian structural equation models with variable selection. We applied a joint Bayesian structural model for the mediator, including a Weibull mixed model for the vital outcome with deprivation as exposure. We selected modifiers via a Monte Carlo model selection procedure. The results suggest that high SES-related indices, such as Yost, Kolak urbanicity (URB), mobility (MOB) and SES dimensions, and concentrated disadvantage index (CDI), all have a significant impact on improved survival. In contrast, area deprivation index (ADI)/Singh, and area level poverty (POV) did not have a major impact. In some cases, the indirect effects have very wide credible intervals, so the total effect is not well estimated despite the estimation of the direct effect. First, it is clear that commonly used indices such as Yost, or CDI both significantly impact the survival experience of Black women diagnosed with epithelial ovarian cancer. In addition, the Kolak dimension indices (URB, MOB, mixed immigrant: MICA and SES) also demonstrate a significant association, depending on the mediator. Mediation effects differ according to the mediator chosen.

A joint Bayesian hierarchical model for estimating SARS-CoV-2 genomic and subgenomic RNA viral dynamics and seroconversion.

Understanding the viral dynamics of and natural immunity to the severe acute respiratory syndrome coronavirus 2 is crucial for devising better therapeutic and prevention strategies for coronavirus disease 2019 (COVID-19). Here, we present a Bayesian hierarchical model that jointly estimates the genomic RNA viral load, the subgenomic RNA (sgRNA) viral load (correlated to active viral replication), and the rate and timing of seroconversion (correlated to presence of antibodies). Our proposed method accounts for the dynamical relationship and correlation structure between the two types of viral load, allows for borrowing of information between viral load and antibody data, and identifies potential correlates of viral load characteristics and propensity for seroconversion. We demonstrate the features of the joint model through application to the COVID-19 post-exposure prophylaxis study and conduct a cross-validation exercise to illustrate the model's ability to impute the sgRNA viral trajectories for people who only had genomic RNA viral load data.

Bayesian multivariate longitudinal model for immune responses to Leishmania: A tick-borne co-infection study.

While many Bayesian state-space models for infectious disease processes focus on population infection dynamics (eg, compartmental models), in this work we examine the evolution of infection processes and the complexities of the immune responses within the host using these techniques. We present a joint Bayesian state-space model to better understand how the immune system contributes to the control of Leishmania infantum infections over the disease course. We use longitudinal molecular diagnostic and clinical data of a cohort of dogs to describe population progression rates and present evidence for important drivers of clinical disease. Among these results, we find evidence for the importance of co-infection in disease progression. We also show that as dogs progress through the infection, parasite load is influenced by their age, ectoparasiticide treatment status, and serology. Furthermore, we present evidence that pathogen load information from an earlier point in time influences its future value and that the size of this effect varies depending on the clinical stage of the dog. In addition to characterizing the processes driving disease progression, we predict individual and aggregate patterns of Canine Leishmaniasis progression. Both our findings and the application to individual-level predictions are of direct clinical relevance, presenting possible opportunities for application in veterinary practice and motivating lines of additional investigation to better understand and predict disease progression. Finally, as an important zoonotic human pathogen, these results may support future efforts to prevent and treat human Leishmaniosis.

Bayesian data synthesis and the utility-risk trade-off for mixed epidemiological data

Much of the microdata used for epidemiological studies contain sensitive measurements on real individuals. As a result, such microdata cannot be published out of privacy concerns, and without public access to these data, any statistical analyses originally published on them are nearly impossible to reproduce. To promote the dissemination of key datasets for analysis without jeopardizing the privacy of individuals, we introduce a cohesive Bayesian framework for the generation of fully synthetic high-dimensional microdatasets of mixed categorical, binary, count, and continuous variables. This process centers around a joint Bayesian model that is simultaneously compatible with all of these data types, enabling the creation of mixed synthetic datasets through posterior predictive sampling. Furthermore, a focal point of epidemiological data analysis is the study of conditional relationships between various exposures and key outcome variables through regression analysis. We design a modified data synthesis strategy to target and preserve these conditional relationships, including both nonlinearities and interactions. The proposed techniques are deployed to create a synthetic version of a confidential dataset containing dozens of health, cognitive, and social measurements on nearly 20,000 North Carolina children.

A joint Bayesian spatiotemporal risk prediction model of COVID-19 incidence, IC admission, and death with application to Sweden

The three closely related COVID-19 outcomes of incidence, intensive care (IC) admission and death, are commonly modelled separately leading to biased estimation of the parameters and relatively poor forecasts. This paper presents a joint spatiotemporal model of the three outcomes based on weekly data that is used for risk prediction and identification of hotspots. The paper applies a pure spatiotemporal model consisting of structured and unstructured spatial and temporal effects and their interaction capturing the effects of the unobserved covariates. The pure spatiotemporal model limits the data requirements to the three outcomes and the population at risk per spatiotemporal unit. The empirical study for the 21 Swedish regions for the period 1 January 2020–4 May 2021 confirms that the joint model predictions outperform the separate model predictions. The fifteen-week-ahead spatiotemporal forecasts (5 May–11 August 2021) show a significant decline in the relative risk of COVID-19 incidence, IC admission, death and number of hotspots.

Optimizing prevalence estimates for a novel pathogen by reducing uncertainty in test characteristics

Emergence of a novel pathogen drives the urgent need for diagnostic tests that can aid in defining disease prevalence. The limitations associated with rapid development and deployment of these tests result in a dilemma: In efforts to optimize prevalence estimates, would tests be better used in the lab to reduce uncertainty in test characteristics or to increase sample size in field studies? Here, we provide a framework to address this question through a joint Bayesian model that simultaneously analyzes lab validation and field survey data, and we define the impact of test allocation on inferences of sensitivity, specificity, and prevalence. In many scenarios, prevalence estimates can be most improved by apportioning additional effort towards validation rather than to the field. The joint model provides superior estimation of prevalence, sensitivity, and specificity, compared with typical analyses that model lab and field data separately, and it can be used to inform sample allocation when testing is limited.

A Joint Analysis for Field Goal Attempts and Percentages of Professional Basketball Players: Bayesian Nonparametric Resource

Understanding shooting patterns among different players is a fundamental problem in basketball game analyses. In this paper, we quantify the shooting pattern via the field goal attempts and percentages over twelve non-overlapping regions around the front court. A joint Bayesian nonparametric mixture model is developed to find latent clusters of players based on their shooting patterns. We apply our proposed model to learn the heterogeneity among selected players from the National Basketball Association (NBA) games over the 2018–2019 regular season and 2019–2020 bubble season. Thirteen clusters are identified for 2018–2019 regular season and seven clusters are identified for 2019–2020 bubble season. We further examine the shooting patterns of players in these clusters and discuss their relation to players’ other available information. The results shed new insights on the effect of NBA COVID bubble and may provide useful guidance for player’s shot selection and team’s in-game and recruiting strategy planning.

Estimation of SARS-CoV-2 antibody prevalence through serological uncertainty and daily incidence.

Serology tests for SARS‐CoV‐2 provide a paradigm for estimating the number of individuals who have had an infection in the past (including cases that are not detected by routine testing, which has varied over the course of the pandemic and between jurisdictions). Such estimation is challenging in cases for which we only have limited serological data and do not take into account the uncertainty of the serology test. In this work, we provide a joint Bayesian model to improve the estimation of the sero‐prevalence (the proportion of the population with SARS‐CoV‐2 antibodies) through integrating multiple sources of data, priors on the sensitivity and specificity of the serological test, and an effective epidemiological dynamics model. We apply our model to the Greater Vancouver area, British Columbia, Canada, with data acquired during the pandemic from the end of January to May 2020. Our estimated sero‐prevalence is consistent with previous literature but with a tighter credible interval.

Clinical Time Delay Distributions of COVID-19 in 2020-2022 in the Republic of Korea: Inferences from a Nationwide Database Analysis.

Epidemiological distributions of the coronavirus disease 2019 (COVID-19), including the intervals from symptom onset to diagnosis, reporting, or death, are important for developing effective disease-control strategies. COVID-19 case data (from 19 January 2020 to 10 January 2022) from a national database maintained by the Korea Disease Control and Prevention Agency and the Central Disease Control Headquarters were analyzed. A joint Bayesian subnational model with partial pooling was used and yielded probability distribution models of key epidemiological distributions in Korea. Serial intervals from before and during the Delta variant’s predominance were estimated. Although the mean symptom-onset-to-report interval was 3.2 days at the national level, it varied across different regions (2.9–4.0 days). Gamma distribution showed the best fit for the onset-to-death interval (with heterogeneity in age, sex, and comorbidities) and the reporting-to-death interval. Log-normal distribution was optimal for ascertaining the onset-to-diagnosis and onset-to-report intervals. Serial interval (days) was shorter before the Delta variant-induced outbreaks than during the Delta variant’s predominance (4.4 vs. 5.2 days), indicating the higher transmission potential of the Delta variant. The identified heterogeneity in region-, age-, sex-, and period-based distributions of the transmission dynamics of COVID-19 will facilitate the development of effective interventions and disease-control strategies.

ET-LDA: Joint Topic Modeling for Aligning Events and their Twitter Feedback

During broadcast events such as the Superbowl, the U.S. Presidential and Primary debates, etc., Twitter has become the de facto platform for crowds to share perspectives and commentaries about them. Given an event and an associated large-scale collection of tweets, there are two fundamental research problems that have been receiving increasing attention in recent years. One is to extract the topics covered by the event and the tweets; the other is to segment the event. So far these problems have been viewed separately and studied in isolation. In this work, we argue that these problems are in fact inter-dependent and should be addressed together. We develop a joint Bayesian model that performs topic modeling and event segmentation in one unified framework. We evaluate the proposed model both quantitatively and qualitatively on two large-scale tweet datasets associated with two events from different domains to show that it improves significantly over baseline models.

Characterizing Social TV Activity Around Televised Events: A Joint Topic Model Approach

Viewers often use social media platforms like Twitter to express their views about televised programs and events like the presidential debate, the Oscars, and the State of the Union speech. Although this promises tremendous opportunities to analyze the feedback on a program or an event using viewer-generated content on social media, there are significant technical challenges to doing so. Specifically, given a televised event and related tweets about this event, we need methods to effectively align these tweets and the corresponding event. In turn, this will raise many questions, such as how to segment the event and how to classify a tweet based on whether it is generally about the entire event or specifically about one particular event segment. In this paper, we propose and develop a novel joint Bayesian model that aligns an event and its related tweets based on the influence of the event’s topics. Our model allows the automated event segmentation and tweet classification concurrently. We present an efficient inference method for this model and a comprehensive evaluation of its effectiveness compared with the state-of-the-art methods. We find that the topics, segments, and alignment provided by our model are significantly more accurate and robust.

Coupling a Generative Model With a Discriminative Learning Framework for Speaker Verification

The task of speaker verification (SV) is to decide whether an utterance is spoken by a target or an imposter speaker. In most studies of SV, a log-likelihood ratio (LLR) score is estimated based on a generative probability model on speaker features, and compared with a threshold for making a decision. However, the generative model usually focuses on individual feature distributions, does not have the discriminative feature selection ability, and is easy to be distracted by nuisance features. The SV, as a hypothesis test, could be formulated as a binary discrimination task where neural network based discriminative learning could be applied. In discriminative learning, the nuisance features could be removed with the help of label supervision. However, discriminative learning pays more attention to classification boundaries, and is prone to overfitting to a training set which may result in bad generalization on a test set. In this paper, we propose a hybrid learning framework, i.e., coupling a joint Bayesian (JB) generative model structure and parameters with a neural discriminative learning framework for SV. In the hybrid framework, a two-branch Siamese neural network is built with dense layers that are coupled with factorized affine transforms as used in the JB model. The LLR score estimation in the JB model is formulated according to the distance metric in the discriminative learning framework. By initializing the two-branch neural network with the generatively learned model parameters of the JB model, we further train the model parameters with the pairwise samples as a binary discrimination task. Moreover, a direct evaluation metric (DEM) in SV based on minimum empirical Bayes risk (EBR) is designed and integrated as an objective function in the discriminative learning. We carried out SV experiments on Speakers in the wild (SITW) and Voxceleb. Experimental results showed that our proposed model improved the performance with a large margin compared with state of the art models for SV.

Inference of COVID-19 epidemiological distributions from Brazilian hospital data.

Knowing COVID-19 epidemiological distributions, such as the time from patient admission to death, is directly relevant to effective primary and secondary care planning, and moreover, the mathematical modelling of the pandemic generally. We determine epidemiological distributions for patients hospitalized with COVID-19 using a large dataset (N = 21 000 − 157 000) from the Brazilian Sistema de Informação de Vigilância Epidemiológica da Gripe database. A joint Bayesian subnational model with partial pooling is used to simultaneously describe the 26 states and one federal district of Brazil, and shows significant variation in the mean of the symptom-onset-to-death time, with ranges between 11.2 and 17.8 days across the different states, and a mean of 15.2 days for Brazil. We find strong evidence in favour of specific probability density function choices: for example, the gamma distribution gives the best fit for onset-to-death and the generalized lognormal for onset-to-hospital-admission. Our results show that epidemiological distributions have considerable geographical variation, and provide the first estimates of these distributions in a low and middle-income setting. At the subnational level, variation in COVID-19 outcome timings are found to be correlated with poverty, deprivation and segregation levels, and weaker correlation is observed for mean age, wealth and urbanicity.

3D shape retrieval based on Laplace operator and joint Bayesian model

Feature analysis plays a significant role in computer vision and computer graphics. In the task of shape retrieval, shape descriptor is indispensable. In recent years, feature extraction based on deep learning becomes very popular, but the design of geometric shape descriptor is still meaningful due to the contained intrinsic information and interpretability. This paper proposes an effective and robust descriptor of 3D models. The descriptor is constructed based on the probability distribution of the normalized eigenfunctions of the Laplace–Beltrami operator on the surface, and a spectrum method for dimensionality reduction. The distance metric of the descriptor space is learned by utilizing the joint Bayesian model, and we introduce a matrix regularization in the training stage to re-estimate the covariance matrix. Finally, we apply the descriptor to 3D shape retrieval on a public benchmark. Experiments show that our method is robust and has good retrieval performance.

Weakly Supervised Object Co-Localization via Sharing Parts Based on a Joint Bayesian Model

Objects in images are characterized by intra-class variation, inter-class diversity, and noisy images. These characteristics pose a challenge to object localization. To address this issue, we present a novel joint Bayesian model for weakly-supervised object localization. The differences compared to previous discriminative methods are evaluated in three aspects: (1) We co-localize the similar object per class through transferring shared parts, which are pooling by modeling object, parts and features within and between-class; (2) Labels are given at class level to provide strong supervision for features and corresponding parts; (3) Noisy images are considered by leveraging a constraint on the detection of shared parts. In addition, our methods are evaluated by extensive experiments. The results indicated outperformance of the state-of-the-art approaches with almost 7% and 1.5% improvements in comparison to the previous methods on PASCAL VOC 2007 6 × 2 and Object Discovery datasets, respectively.

Multifractal Analysis of Multivariate Images Using Gamma Markov Random Field Priors

Texture characterization of natural images using the mathematical framework of multifractal analysis (MFA) enables the study of the fluctuations in the regularity of image intensity. Although successfully applied in various contexts, the use of MFA has so far been limited to the independent analysis of a single image, while the data available in applications are increasingly multivariate. This paper addresses this limitation and proposes a joint Bayesian model and associated estimation procedure for multifractal parameters of multivariate images. It builds on a recently introduced generic statistical model that enabled the Bayesian estimation of multifractal parameters for a single image and relies on the following original key contributions: First, we develop a novel Fourier domain statistical model for a single image that permits the use of a likelihood that is separable in the multifractal parameters via data augmentation. Second, a joint Bayesian model for multivariate images is formulated in which prior models based on gamma Markov random fields encode the assumption of the smooth evolution of multifractal parameters between the image components. The design of the likelihood and of conjugate prior models is such that exploitation of the conjugacy between the likelihood and prior models enables an efficient estimation procedure that can handle a large number of data components. Numerical simulations conducted using sequences of multifractal images demonstrate that the proposed procedure significantly outperforms previous univariate benchmark formulations at a competitive computational cost.

Joint Bayesian guided metric learning for end-to-end face verification

Abstract In this work, we address the problem of face verification, namely determining whether a pair of face images belongs to the same or different subjects. Previous works often consider solving the problem of face verification in two steps: feature extraction and face recognition, resulting in a fragmented procedure. We argue that these techniques, although working well, fail to explicitly exploit a full end-to-end framework for face verification, which has received much attention and achieved significant improvements recently. In this paper, we propose a novel Joint Bayesian guided metric learning technique for dealing with the face verification task, which well integrates the above two steps of face verification into an end-to-end convolutional neural network (CNN) architecture. In the training stage, an initial neural network, which has the similar architecture with GoogLeNet CNN model, is firstly pre-trained by optimizing classification-based objective functions on the publicly available CASIA WebFace database. Based on constructed face pairs dataset from CASIA WebFace and LFW datasets, we then fine-tune the whole network parameters under the guide of the learned knowledge, which is obtained from the highly successful Joint Bayesian model. This guided learning procedure, which can also be seen as a metric learning technique, can further update network parameters for discriminating face pairs. In the testing process, the outputs by this unified network are discriminated with a threshold value to produce the ultimate prediction for the face verification task. Comprehensive evaluations over the LFW dataset well demonstrate the encouraging face verification performance of our proposed framework.