A5022154435- Statistical Methods and Inference
- Advanced Statistical Methods and Models
- Face and Expression Recognition
- Neural Networks and Applications
- Geriatric Care and Nursing Homes
- Bayesian Methods and Mixture Models
- Financial Risk and Volatility Modeling
- Chronic Disease Management Strategies
- Frailty in Older Adults
- Imbalanced Data Classification Techniques
- Anomaly Detection Techniques and Applications
- Computational Drug Discovery Methods
- Statistical Methods and Bayesian Inference
- Complex Systems and Time Series Analysis
- Dementia and Cognitive Impairment Research
- RNA Research and Splicing
- MicroRNA in disease regulation
- Protein Structure and Dynamics
- Genetic Associations and Epidemiology
- Gene expression and cancer classification
- Sports Dynamics and Biomechanics
- Statistical Distribution Estimation and Applications
- Machine Learning and Algorithms
- Image Retrieval and Classification Techniques
- Machine Learning and Data Classification
University of Waterloo
2013-2024
Chinese Center For Disease Control and Prevention
2021-2023
Center for Disease Control
2021
Acadia University
2017
Beijing University of Chinese Medicine
2017
Jinan University
2017
Royal Bank of Canada
2017
China Astronaut Research and Training Center
2012
Dalhousie University
2010
Stanford University
2003
Abstract Recent studies indicate important roles for long noncoding RNAs (lncRNAs) as essential regulators of myogenesis and adult skeletal muscle regeneration. However, the specific lncRNAs in myogenic differentiation stem cells are still largely unknown. Here we identify a lncRNA that is specifically enriched (myogenesis-associated lncRNA, short, lnc-mg). In mice, conditional knockout lnc-mg results atrophy loss muscular endurance during exercise. Alternatively, muscle-specific...
The Netflix prize has rejuvenated a widespread interest in the matrix factorization approach for collaborative filtering. We describe simple algorithm incorporating content information directly into this approach. present experimental evidence using recipe data to show that not only improves recommendation accuracy but also provides useful insights about contents themselves are otherwise unavailable.
For similar evaluation tasks, the area under receiver operating characteristic curve (AUC) is often used by researchers in machine learning, whereas average precision (AP) more information retrieval community. We establish some results to explain why this case. Specifically, we show that, when both AUC and AP are rescaled lie [0,1], approximately times initial of system.
In Bayesian statistics, the choice of prior distribution is often controversial. Different rules for selecting priors have been suggested in literature, which, sometimes, produce that are difficult students to understand intuitively. this article, we use a simple heuristic illustrate rather counter-intuitive fact flat not necessarily non-informative; and non-informative flat.
The need to identify a few important variables that affect certain outcome of interest commonly arises in various industrial engineering applications. genetic algorithm (GA) appears be natural tool for solving such problem. In this article we first demonstrate the GA is actually not particularly effective variable selection tool, and then propose very simple modification. Our idea run number GAs parallel without allowing each fully converge, consolidate information from all individual end....
Targeting older clients for rehabilitation is a clinical challenge and research priority. We investigate the potential of machine learning algorithms - Support Vector Machine (SVM) K-Nearest Neighbors (KNN) to guide planning home care clients.This study secondary analysis data on 24,724 longer-term from eight programs in Ontario. Data were collected with RAI-HC assessment system, which Activities Daily Living Clinical Assessment Protocol (ADLCAP) used identify potential. For purposes, client...
Lu Xin is PhD student, Department of Statistics and Actuarial Science, University Waterloo, Ontario N2L 3G1, Canada (E-mail: lxin@uwaterloo.ca). Mu Zhu Associate Professor, m3zhu@uwaterloo.ca).
For professional basketball, finding valuable and suitable players is the key to building a winning team. To deal with such challenges, basketball managers, scouts coaches are increasingly turning analytics. Objective evaluation of teams has always been top goal Typical statistical analytics mainly focuses on box score developed various metrics. In spite more advanced methods, metrics built upon statistics provide limited information about how interact each other. Two similar scores may...
In high-dimensional classification problems, one is often interested in finding a few important discriminant directions order to reduce the dimensionality. Fisher's linear analysis (LDA) commonly used method. Although LDA guaranteed find best when each class has Gaussian density with common covariance matrix, it can fail if densities are more general. Using likelihood-based interpretation of criterion, we develop general method for without assuming belong any particular parametric family. We...
The area under the receiver operating characteristic curve (AUC) is frequently used as a performance measure for medical tests. It threshold-free that independent of disease prevalence rate. We evaluate utility AUC against an alternate called average positive predictive value (AP), in setting many screening programs where has low rate.We define two measures using common notation system and show both can be expressed weighted density function diseased subjects. weights AP include some form,...
To examine regional variation in service provision and identify the client characteristics associated with occupational therapy (OT) physiotherapy (PT) services for older adults Ontario Home Care System.Secondary analyses of a provincial database containing comprehensive assessments (RAI-HC) linked utilization data from every long-stay home care system between 2005 2010 (n = 299 262). Hierarchical logistic regression models were used to model dependent variables OT PT use within 90 d initial...
Abstract Using Cox regression as the main platform, we study ensemble approach for variable selection. We use a popular real-data example well simulated data with various censoring levels to illustrate usefulness of approach, and nature these ensembles in terms their strength diversity. By relating characteristics ensemble's selection accuracy, provide useful insights how choose among different strategies, guidelines thinking about design more effective ensembles. Keywords: bootstrapCox...
Generative moment matching networks (GMMNs) are introduced for generating approximate quasi-random samples from multivariate models with any underlying copula to compute estimates variance reduction. So far, sampling distributions required a careful design, exploiting specific properties (such as conditional distributions) of the implied parametric or quasi-Monte Carlo (QMC) point set, and was only tractable small number models. Using GMMNs allows one construct much larger variety without...
We study a general class of statistical detection problems where the underlying objective is to detect items belonging rare from very large database. propose computationally efficient method achieve this goal. Our consists two steps. In first step we estimate density function alone with an adaptive bandwidth kernel estimator. The choice inspired by ancient Chinese board game known today as Go. second adjust locally depending on background nearby. show that amount adjustment needed in...
Microgravity has been previously demonstrated to induce skeletal muscle atrophy, loss of force and disorders in myogenesis metabolism. Current pharmacological strategies exhibit poor efficacy. Bu Zhong Yi Qi decoction (BZ) is a well‑known traditional Chinese medicine used for myasthenia gravis. In the present study, its effect on unloading induced atrophy was investigated. The mousetail suspension model simulate weightlessness atrophy. results indicated that BZ could significantly protect...
Resources for home care rehabilitation are limited, and many clients who could benefit do not receive therapy. The interRAI Contact Assessment (CA) is a new screening instrument comprised of subset Home Care (HC) items, designed to be used as preliminary assessment identify which potential should referred full assessment, or services such rehabilitation. We investigated client characteristics most relevant in predicting use the HC assessment. applied two algorithms from machine learning data...