Article
Statistics & Probability
Arkaprava Roy, Brian J. Reich, Joseph Guinness, Russell T. Shinohara, Ana-Maria Staicu
Summary: This study introduces a novel approach for sparse signal detection on a spatial domain by modeling continuous signals as the product of independent Gaussian processes. The proposed method achieves better control over signal smoothness and sparsity using PING processes, resulting in improved estimation accuracy for image regressions.
JOURNAL OF COMPUTATIONAL AND GRAPHICAL STATISTICS
(2021)
Article
Biochemistry & Molecular Biology
Leo A. Featherstone, Sebastian Duchene, Timothy G. Vaughan
Summary: Despite its increasing role in understanding infectious disease transmission, phylodynamics lacks clarity on ideal data and optimal sampling. This study introduces a method to visualize and quantify the impact of pathogen genome sequence and sampling times on phylodynamic inference. By applying the method to simulated and real-world data, the study provides insights and guidelines for maximizing the use of sequence data in phylodynamic analyses. The continued research on phylodynamic data and inference is crucial for targeted and efficient responses to infectious disease threats.
MOLECULAR BIOLOGY AND EVOLUTION
(2023)
Article
Statistics & Probability
Philip A. White, Durban G. Keeler, Summer Rupper
Summary: The article introduces a new method to estimate snow density in Antarctica, as well as analyzes the trend of water accumulation. Through interpolation, snow density can be estimated in regions where snow cores have not been drilled, and it has been found that water accumulation has mainly decreased in recent decades.
ANNALS OF APPLIED STATISTICS
(2021)
Article
Neurosciences
Gargi Majumdar, Fahd Yazin, Arpan Banerjee, Dipanjan Roy
Summary: This study explores the role of uncertainty estimation in governing affective dynamics, showing that emotions naturally arise due to ongoing uncertainty estimations about future outcomes. The lateral orbitofrontal cortex (lOFC) tracks the temporal fluctuations of this uncertainty and is predictive of anxiety predisposition.
Article
Engineering, Electrical & Electronic
Zhe Wang, Yongxiong Wang, Chuanfei Hu, Zhong Yin, Yu Song
Summary: This study proposes a transformer-based model to hierarchically learn the discriminative spatial information from electrode level to brain-region-level in emotion recognition. Experimental results demonstrate that the proposed model achieves outstanding performance and can emphasize discriminative spatial information from specific brain regions.
IEEE SENSORS JOURNAL
(2022)
Article
Ecology
Leo A. Featherstone, Francesca Di Giallonardo, Edward C. Holmes, Timothy G. Vaughan, Sebastian Duchene
Summary: The article discusses incorporating un-sequenced case occurrence data alongside sequenced data in Phylodynamic analysis, demonstrating through simulations that this approach can eliminate bias in estimates of the basic reproductive number due to misspecification of the sampling process. Additionally, it emphasizes that occurrence data are a valuable source of information for future Phylodynamic analyses.
METHODS IN ECOLOGY AND EVOLUTION
(2021)
Article
Engineering, Industrial
Linhan Ouyang, Chanseok Park, Yan Ma, Yizhong Ma, Min Wang
Summary: This paper introduces a Bayesian hierarchical modelling approach and SUR model for process optimisation, aiming to improve efficiency through model selection and estimation. Additionally, a two-stage optimisation strategy considering practitioners' preference information is proposed to find the best settings and compare them with an ideal solution method.
INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH
(2021)
Article
Computer Science, Artificial Intelligence
Tingting Zhu, Pengfei Bi, Mengmeng Wang
Summary: Personalization is a prominent marketing strategy for retailers both online and offline. However, the sparse customer behavioral data makes it challenging to calculate personalized preferences and predict behaviors. To address this, a Hierarchical Nonparametric Bayesian method (HNB) is proposed to model personalized preferences based on latent groups. HNB utilizes a hierarchical and nonparametric structure to analyze preferences and accurately identify latent groups from customer purchase history. Experimental results demonstrate the automatic calculation of preferences and latent groups, understanding the generative mechanism of personalized preference, and accurate prediction of personalized purchases.
KNOWLEDGE-BASED SYSTEMS
(2023)
Article
Ecology
Jeffrey W. Doser, Andrew O. Finley, Sudipto Banerjee
Summary: Determining the spatial distributions of species and communities is important in ecology and conservation efforts. We developed a spatial factor multi-species occupancy model to explicitly account for species correlations, imperfect detection, and spatial autocorrelation. Ignoring these complexities leads to inferior model predictive performance, and our proposed model had the highest predictive performance among the alternative models.
Article
Mathematics, Applied
Thierno Souleymane Barry, Oscar Ngesa, Jeremiah Kimani Kiingati, Nelson Owuor Onyango, Aurise Niyoyunguruza, Alexis Habineza, Henry Mwambi, Henri Bello Fika
Summary: Anemia and malaria are the leading causes of morbidity and mortality among children under five years old in sub-Saharan Africa. This study aims to estimate the spatial linear correlation between anemia and malaria, and investigate the factors affecting morbidity in Guinea. The findings show high prevalences of anemia and malaria in children under five years old in Guinea, and significant associations between each disease and various demographic factors.
Article
Computer Science, Interdisciplinary Applications
Diogo Ferrari
Summary: The existence of latent clusters with different responses to treatment is a significant concern in scientific research. This article discusses the implementation of a novel hierarchical Dirichlet process approach using the R package hdpGLM. The methods provided in the package make it easier for researchers to investigate heterogeneity in treatment effects and identify clusters of subjects with differential effects.
JOURNAL OF STATISTICAL SOFTWARE
(2023)
Article
Mathematics
Junbo Zhang, Daoji Li, Yingzhi Xia, Qifeng Liao
Summary: This paper proposes a new two-step approach to estimate 1-km-resolution PM2.5 concentrations in Shanghai using satellite data. The approach refines AOD data to a higher resolution using a Bayesian retrieval method and then uses a hierarchical Gaussian process model to estimate PM2.5 concentrations. The results show accurate predictive performance of the proposed approach.
Article
Automation & Control Systems
Marta Catalano, Pierpaolo De Blasi, Antonio Lijoi, Igor Prunster
Summary: Bayesian hierarchical models are powerful tools for learning common latent features across multiple data sources. This study establishes theoretical guarantees for recovering the true data generating process in the Hierarchical Dirichlet Process (HDP) or a generalization of the HDP. The posterior contraction rates are affected by the relationship between sample sizes.
JOURNAL OF MACHINE LEARNING RESEARCH
(2022)
Article
Biology
Weiji Su, Xia Wang, Rhonda D. Szczesniak
Summary: This study introduces a novel joint model to investigate the association between lung function decline and recurrent pulmonary exacerbation events more accurately. By using Gaussian processes and flexible link functions, the model improves fit to the data and characterizes the relationship between accelerated lung function decline and increased odds of experiencing another PE.
Article
Engineering, Civil
Alvaro Ossandon, Balaji Rajagopalan, William Kleiber
Summary: The semi-Bayesian hierarchical modeling framework combines generalized extreme value distribution and Gaussian multivariate process to analyze precipitation extremes over a large domain. By conducting space-time frequency analysis of seasonal maximum precipitation, the model captures historical variability well and has wide applications in natural resources and infrastructure management.
JOURNAL OF HYDROLOGY
(2021)
Article
Mathematical & Computational Biology
Tianyu Zhan, Alan Hartford, Jian Kang, Walter Offen
Summary: This article evaluates the performance of two existing constrained methods and proposes a deep learning enhanced optimization framework. By using feedforward neural networks to approximate the objective function and utilizing gradient information for optimization, our method achieves a better balance between robustness and time efficiency.
STATISTICS IN BIOPHARMACEUTICAL RESEARCH
(2022)
Article
Biology
Cui Guo, Jian Kang, Timothy D. Johnson
Summary: Image-on-image regression analysis is challenging due to high dimensionality and complex spatial dependence. The proposed model effectively captures spatial dependence among image outcomes and predictors, achieving better prediction accuracy and dimension reduction. By incorporating spatial Bayesian latent factor model and Gaussian process priors, the method demonstrates improved performance in predicting task-related contrast maps in multimodal image data.
Article
Biology
Kevin He, Ji Zhu, Jian Kang, Yi Li
Summary: Analyzing the national transplant database, a blockwise steepest ascent procedure is proposed to fit a time-varying effect model, with a Wald statistic used to test if effects indeed vary over time. The utility of the method is evaluated through simulations and applied to analyze national kidney transplant data, detecting time-varying effects of various risk factors.
Article
Public, Environmental & Occupational Health
Chao Yuan, Hongtao Yang, Siyuan Zheng, Xiangyu Sun, Xiaochi Chen, Yuntao Chen, Jian Kang, Moubin Liu, Shuguo Zheng
Summary: In this study, the distributions of dental splatters and the effectiveness of corresponding control measures were evaluated using high-speed videography and laser diffraction. The majority of dental splatters were found to be small droplets (<50 µm). The combination of high-volume evacuation and suction air purifier was able to clear away most of the droplets and aerosols.
INFECTION CONTROL AND HOSPITAL EPIDEMIOLOGY
(2023)
Article
Biology
Yize Zhao, Ben Wu, Jian Kang
Summary: In functional neuroimaging studies, identifying predictive imaging markers and intermodality interactions is crucial for understanding brain activity. This paper presents a unified Bayesian prior model that simultaneously identifies main effect features and intermodality interactions using intermediate selection status, improving posterior inference accuracy and enhancing biological plausibility. Extensive simulations and application to real data demonstrate the superiority of this approach.
Article
Respiratory System
Sarah L. Finnegan, Olivia K. Harrison, Sara Booth, Andrea Dennis, Martyn Ezra, Catherine J. Harmer, Mari Herigstad, Bryan Guillaume, Thomas E. Nichols, Najib M. Rahman, Andrea Reinecke, Olivier Renaud, Kyle T. S. Pattinson
Summary: The study found that D-cycloserine did not improve the efficacy of pulmonary rehabilitation in treating chronic breathlessness. However, it did interact with breathlessness anxiety. Therefore, a phase 3 clinical trial of D-cycloserine may not be worthwhile.
Article
Gastroenterology & Hepatology
Adriana Roca-Fernandez, Rajarshi Banerjee, Helena Thomaides-Brears, Alison Telford, Arun Sanyal, Stefan Neubauer, Thomas E. Nichols, Betty Raman, Celeste McCracken, Steffen E. Petersen, Ntobeko A. B. Ntusi, Daniel J. Cuthbertson, Michele Lai, Andrea Dennis, Amitava Banerjee
Summary: This study found that early signs of liver disease, measured by cT1, were associated with an increased risk of cardiovascular disease. Liver disease activity (cT1) was associated with major CVD events, CVD hospitalisation, and all-cause mortality, while liver fat (PDFF) was not associated.
JOURNAL OF HEPATOLOGY
(2023)
Article
Radiology, Nuclear Medicine & Medical Imaging
Ahmed M. M. Salih, Esmeralda Ruiz Pujadas, Victor M. Campello, Celeste McCracken, Nicholas C. C. Harvey, Stefan Neubauer, Karim Lekadir, Thomas E. E. Nichols, Steffen E. E. Petersen, Zahra Raisi-Estabragh
Summary: This study estimates the biological age of different cardiac regions using magnetic resonance imaging radiomics phenotypes and investigates the determinants of aging in those regions. The results show associations between age gap and factors like visceral adiposity, mental health, dental problems, and bone mineral density.
JOURNAL OF MAGNETIC RESONANCE IMAGING
(2023)
Article
Statistics & Probability
Jinyuan Chang, Jing He, Jian Kang, Mingcong Wu
Summary: This article proposes rigorous statistical testing procedures for making inferences on the complex dependence of multimodal imaging data. The proposed methods address three hypothesis testing problems and include a global testing procedure and a multiple testing procedure controlling the false discovery rate. Extensive simulations and analysis of task fMRI contrast maps validate the accuracy and effectiveness of the proposed methods.
JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION
(2023)
Letter
Multidisciplinary Sciences
Brenden Tervo-Clemmens, Scott Marek, Roselyne J. Chauvin, Andrew N. Van, Benjamin P. Kay, Timothy O. Laumann, Wesley K. Thompson, Thomas E. Nichols, B. T. Thomas Yeo, Deanna M. Barch, Beatriz Luna, Damien A. Fair, Nico U. F. Dosenbach
Article
Multidisciplinary Sciences
Anya C. Topiwala, Thomas Nichols, Logan Z. J. Williams, Emma Robinson, Fidel P. Alfaro-Almagro, Bernd L. Taschler, Chaoyue Wang, Christopher J. Nelson, Karla M. Miller, Veryan Codd, Nilesh Samani, Stephen Smith
Summary: Telomeres form protective caps at the ends of chromosomes and their attrition is linked to biological aging. Short telomeres are associated with increased risk of neurological and psychiatric disorders, including dementia. The relationship between telomere length and neuroimaging markers is not well-defined.
Article
Biology
Yuan Yang, Ziyang Pan, Jian Kang, Chad Brummett, Yi Li
Summary: Varying coefficient models with zero regions are proposed in this study. The new modeling approach allows for variable selection, detects zero regions, obtains point estimates of varying coefficients, and constructs sparse confidence intervals accommodating zero regions. The asymptotic properties of the estimator are proven for statistical inference. Simulation results show that the proposed sparse confidence intervals have desired coverage probability. The method is applied to analyze a large-scale preoperative opioid study.
Article
Mathematical & Computational Biology
Farhad Hatami, Alex Ocampo, Gordon Graham, Thomas E. Nichols, Habib Ganjgahi
Summary: In this article, an optimization technique for fitting continuous time Markov models (CTMM) in the presence of covariates is proposed. This technique combines a stochastic gradient descent algorithm with differentiation of the matrix exponential using a Pade approximation, making it feasible to fit large scale data. Two methods for computing standard errors are presented, one utilizing the Pade expansion and the other using power series expansion of the matrix exponential. Simulation results show improved performance compared to existing CTMM methods, and the method is demonstrated on a large-scale multiple sclerosis NO.MS dataset.
Article
Neurosciences
Gina-Isabelle Henze, Julian Konzok, Brigitte M. Kudielka, Stefan Wuest, Thomas E. Nichols, Ludwig Kreuzpointner
Summary: This study investigated the relationship between neural measures of limbic structures and hypothalamic pituitary adrenal axis responses to acute stress exposure in healthy young adults. The findings suggest that limbic volume and thickness measures are associated with acute cortisol stress responses, providing new insights into the involvement of striato-limbic structures in psychosocial stress processing.
EUROPEAN JOURNAL OF NEUROSCIENCE
(2023)
Article
Statistics & Probability
Daiwei Zhang, Lexin Li, Chandra Sripada, Jian Kang
Summary: A novel non-parametric approach is proposed to delineate associations between images and covariates using deep neural networks in the framework of spatially varying coefficient models. The method incorporates spatial smoothness, handles subject heterogeneity, and provides straightforward interpretations. It is also highly flexible and accurate, making it ideal for capturing complex association patterns.
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY
(2023)