A5100438701- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Statistical Distribution Estimation and Applications
- Statistical Methods in Clinical Trials
- Advanced Statistical Methods and Models
- Nonlinear Differential Equations Analysis
- Advanced Causal Inference Techniques
- Genetic Associations and Epidemiology
- Fractional Differential Equations Solutions
- Immunotherapy and Immune Responses
- Differential Equations and Numerical Methods
- Adipose Tissue and Metabolism
- Glioma Diagnosis and Treatment
- CAR-T cell therapy research
- Cancer Immunotherapy and Biomarkers
- Liver Disease Diagnosis and Treatment
- Gene expression and cancer classification
- Advanced Statistical Process Monitoring
- Spatial and Panel Data Analysis
- Neural Networks and Applications
- Face and Expression Recognition
- Neuroendocrine Tumor Research Advances
- Adipokines, Inflammation, and Metabolic Diseases
- Differential Equations and Boundary Problems
University of California, Los Angeles
2016-2025
Eisai (United States)
2023
Yangzhou University
2013-2022
University of Southern California
2008-2021
UCLA Health
2021
UCLA Jonsson Comprehensive Cancer Center
2016
Shandong University of Science and Technology
2015
Nanjing Institute of Technology
2013
University of Illinois Urbana-Champaign
2005
Howard Hughes Medical Institute
2003
Summary In this article we study a joint model for longitudinal measurements and competing risks survival data. Our provides flexible approach to handle possible nonignorable missing data in the due dropout. It is also an extension of previous models with single failure type, offering way informatively censored events as risk. consists linear mixed effects submodel outcome proportional cause‐specific hazards frailty ( Prentice et al., 1978 , Biometrics 34, 541–554) data, linked together by...
Receiver operating characteristic (ROC) curve is an effective and widely used method for evaluating the discriminating power of a diagnostic test or statistical model. As useful method, wealth literature about its theories computation methods has been established. The research on ROC curves, however, focused mainly cross-sectional design. Very little estimating curves their summary statistics, especially significance testing, conducted repeated measures Due to complexity standard error...
Although immunotherapeutic strategies are emerging as adjunctive treatments for cancer, sensitive methods of monitoring the immune response after treatment remain to be established. We used a novel next-generation sequencing approach determine whether quantitative assessments tumor-infiltrating lymphocyte (TIL) content and degree overlap T-cell receptor (TCR) sequences in brain tumors peripheral blood were predictors overall survival glioblastoma patients treated with autologous tumor...
Metagenomics data have been growing rapidly due to the advances in NGS technologies. One goal of human microbial studies is detect abundance differences across clinical conditions. Besides small sample size and high dimension, metagenomics are usually represented as compositions (proportions) with a large number zeros skewed distribution. Efficient tools for handling such compositional need be developed. We propose zero-inflated beta regression approach (ZIBSeq) identifying differentially...
Abstract Joint analysis of longitudinal measurements and survival data has received much attention in recent years. However, previous work primarily focused on a single failure type for the event time. In this paper we consider joint modelling repeated competing risks time to allow more than one distinct endpoint which occurs frequently clinical trials. Our model uses latent random variables common covariates link together sub‐models data, respectively. An EM‐based algorithm is derived...
Existing joint models for longitudinal and survival data are not applicable ordinal outcomes with possible non-ignorable missing values caused by multiple reasons. We propose a model measurements competing risks failure time data, in which partial proportional odds the outcome is linked to event times latent random variables. At endpoint, our adopts framework types at same time. The model, as an extension of popular outcomes, more flexible provides tool test assumption. use likelihood...
This article studies a general joint model for longitudinal measurements and competing risks survival data. The consists of linear mixed effects sub-model the outcome, proportional cause-specific hazards frailty data, regression variance–covariance matrix multivariate latent random based on modified Cholesky decomposition. provides useful approach to adjust non-ignorable missing data due dropout enables analysis outcome with informative censoring intermittently measured time-dependent...
Some interesting recent studies have shown that neural network models are useful alternatives in modeling survival data when the assumptions of a classical parametric or semiparametric model such as Cox (1972) seriously violated. However, to best our knowledge, plausibility adapting emerging extreme learning machine (ELM) algorithm for single‐hidden‐layer feedforward networks analysis has not been explored. In this paper, we present kernel ELM regularized by an L 0 ‐based broken adaptive...
Pharmacological treatments that can concomitantly address cigarette smoking and heavy drinking stand to improve health care delivery for these highly prevalent co-occurring conditions. This superiority trial compared the combination of varenicline naltrexone against alone cessation reduction among heavy-drinking smokers.This was a phase 2 randomized double-blind clinical trial. Participants (N=165) who were daily smokers drank heavily received either mg/day plus 50 or matched placebo pills...
Abstract In this article a control percentile test, chi-squared and Kolmogorov-type test are proposed for comparing two distributions from incomplete survival data. These tests obtained by examining vertical shift comparison function at single point, finite number of points, an entire set points on interval. The methods also have applications in receiver operating characteristic (ROC) analysis, which has been widely used such diverse fields as signal detection theory, psychology,...
Abstract BACKGROUND: Sarcomatoid features in renal cell carcinoma may represent an aggressive subclone arising from the primary tumor. The patterns of metastases for these tumors were evaluated to determine if sarcomatoid retained at metastasis and whether percentage tumor influenced spread. METHODS: All patients with found nephrectomy synchronous or metachronous resection evaluated. histology, grade, metastatic site recorded. association between features, pattern was RESULTS: Thirty‐two...
Most hazard regression models in survival analysis specify a given functional form to describe the influence of covariates on rate. For instance, Cox's model assumes that act multiplicatively rate, and Aalen's additive risk stipulates have linear effect In this paper we study fully nonparametric which makes no assumption association between rate covariates. We propose class estimators for conditional function, cumulative function their large sample properties. When size data set is...
Abstract Existing methods for joint modeling of longitudinal measurements and survival data can be highly influenced by outliers in the outcome. We propose a model analysis competing risks failure time which is robust presence outlying observations during follow‐up. Our consists linear mixed effects sub‐model outcome proportional cause‐specific hazards frailty data, linked together latent random effects. Instead usual normality assumption measurement errors sub‐model, we adopt t...
Variable selections for regression with high-dimensional big data have found many applications in bioinformatics and computational biology. One appealing approach is the L0 regularized which penalizes number of nonzero features model directly. However, it well known that optimization NP-hard computationally challenging. In this paper, we propose efficient EM (L0EM) dual L0EM (DL0EM) algorithms directly approximate problem. While large sample size, DL0EM (n ≪ m) data. They also provide a...
Abstract Context Little is known about presenting clinical characteristics, tumor biology, and surgical morbidity of Cushing’s disease (CD) with aging. Objective Using a large multi-institutional dataset, we assessed diagnostic prognostic significance age in CD through differences presentation, laboratory results, postoperative outcomes. Design Data from the Registry Adenomas Pituitary Related Disorders (RAPID) were reviewed for patients treated transsphenoidal resection at 11 centers...
<p>Table S1 shows the gene expression in pancreas as profiled with Mouse Neurotransmitter Receptors RT Profile PCR Array</p>
<p>Table S2 shows the gene expression in pancreas as profiled with Mouse Dopamine and Serotonin pathway RT Profiler PCR Array</p>