A5081246207- Bayesian Methods and Mixture Models
- Machine Learning in Healthcare
- Venous Thromboembolism Diagnosis and Management
- Domain Adaptation and Few-Shot Learning
- Atrial Fibrillation Management and Outcomes
- Central Venous Catheters and Hemodialysis
- Medication Adherence and Compliance
- Statistical Methods and Bayesian Inference
- Dietary Effects on Health
- Reinforcement Learning in Robotics
- Anomaly Detection Techniques and Applications
- Topic Modeling
- Allergic Rhinitis and Sensitization
- Genetics and Plant Breeding
- Gene expression and cancer classification
- Clinical Nutrition and Gastroenterology
- Nutritional Studies and Diet
- Salmonella and Campylobacter epidemiology
- Culinary Culture and Tourism
- Ocular Surface and Contact Lens
- Organ Transplantation Techniques and Outcomes
- Advanced Causal Inference Techniques
- Innovation Diffusion and Forecasting
- Neural Networks and Applications
- Escherichia coli research studies
University of Michigan
2018-2024
University of North Carolina at Chapel Hill
2017-2019
Xijing Hospital
2017
Air Force Medical University
2017
A one-health perspective may provide new and actionable information about Escherichia coli transmission. E. colonizes a broad range of vertebrates, including humans food-production animals, is leading cause bladder, kidney, bloodstream infections in humans. Substantial evidence supports foodborne transmission pathogenic strains from food animals to However, the relative contribution zoonotic (FZEC) human extraintestinal disease burden distinguishing characteristics such remain undefined....
To determine whether improvement in the severity of dry eye disease (DED) symptoms correlates with anxiety and depression.This prospective interventional case series recruited 45 adults evidence DED. Patients were administered University North Carolina Dry Eye Management Scale (DEMS), Generalized Anxiety Disorder 7-item scale (GAD-7), Personal Health Questionnaire Depression (PHQ-8) to evaluate DED symptoms, anxiety, depression, respectively. Standard care treatment was provided for patients...
Determining causes of deaths (CODs) occurred outside civil registration and vital statistics systems is challenging. A technique called verbal autopsy (VA) widely adopted to gather information on in practice. VA consists interviewing relatives a deceased person about symptoms the period leading death, often resulting multivariate binary responses. While statistical methods have been devised for estimating cause-specific mortality fractions (CSMFs) study population, continued expansion new...
Every effort should be made to optimize surgical techniques and minimize potential morbidity rates associated with live donor operations. Advances in a minimally invasive approach by robotic surgery nephrectomy have raised the possibility of applying this technique bowel resections for intestinal transplantation.We report first 5 consecutive cases robotic-assisted ileal segmentectomy. We describe technical aspects procedure, discuss rationale considering option, evaluate advantages...
This paper is concerned with using multivariate binary observations to estimate the probabilities of unobserved classes scientific meanings. We focus on setting where additional information about sample similarities available and represented by a rooted weighted tree. Every leaf in given tree contains multiple samples. Shorter distances over between leaves indicate priori higher similarity class probability vectors. propose novel data integrative extension classical latent models...
Dietary patterns synthesize multiple related diet components, which can be used by nutrition researchers to examine diet-disease relationships. Latent class models (LCMs) have been derive dietary from intake assessment, where each profile represents the probabilities of exposure a set components. However, LCM-derived exhibit strong similarities, or weak separation, resulting in numerical and inferential instabilities that challenge scientific interpretation. This issue is exacerbated...
Determining causes of deaths (COD) occurred outside civil registration and vital statistics systems is challenging. A technique called verbal autopsy (VA) widely adopted to gather information on in practice. VA consists interviewing relatives a deceased person about symptoms the period leading death, often resulting multivariate binary responses. While statistical methods have been devised for estimating cause-specific mortality fractions (CSMFs) study population, continued expansion new...
Traditional applications of latent class models (LCMs) often focus on scenarios where a set unobserved classes are well-defined and easily distinguishable. However, in numerous real-world applications, these weakly separated difficult to distinguish, creating significant numerical challenges. To address issues, we have developed an R package ddtlcm that provides comprehensive analysis visualization tools designed enhance the robustness interpretability LCMs presence weak separation,...
We consider offline reinforcement learning (RL) methods in possibly nonstationary environments. Many existing RL algorithms the literature rely on stationarity assumption that requires system transition and reward function to be constant over time. However, is restrictive practice likely violated a number of applications, including traffic signal control, robotics mobile health. In this paper, we develop consistent procedure test nonstationarity optimal Q-function based pre-collected...
This paper studies reinforcement learning (RL) in doubly inhomogeneous environments under temporal non-stationarity and subject heterogeneity. In a number of applications, it is commonplace to encounter datasets generated by system dynamics that may change over time population, challenging high-quality sequential decision making. Nonetheless, most existing RL solutions require either stationarity or homogeneity, which would result sub-optimal policies if both assumptions were violated. To...
S ummary This paper is concerned with using multivariate binary observations to estimate the probabilities of unobserved classes scientific meanings. We focus on setting where additional information about sample similarities available and represented by a rooted weighted tree. Every leaf in given tree contains multiple samples. Shorter distances over between leaves indicate priori higher similarity class probability vectors. propose novel data integrative extension classical latent models...
S ummary Determining causes of deaths (COD) occurred outside civil registration and vital statistics systems is challenging. A technique called verbal autopsy (VA) widely adopted to gather information on in practice. VA consists interviewing relatives a deceased person about symptoms the period leading death, often resulting multivariate binary responses. While statistical methods have been devised for estimating cause-specific mortality fractions (CSMFs) study population, continued...