A5001187649- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Statistical Distribution Estimation and Applications
- Bayesian Methods and Mixture Models
- Insurance, Mortality, Demography, Risk Management
- Cardiac electrophysiology and arrhythmias
- Genetic and phenotypic traits in livestock
- Total Knee Arthroplasty Outcomes
- Health Systems, Economic Evaluations, Quality of Life
- Advanced Statistical Methods and Models
- Advanced Causal Inference Techniques
- Acute Myocardial Infarction Research
- Knee injuries and reconstruction techniques
- Plant Pathogens and Resistance
- Genetics and Plant Breeding
- Health disparities and outcomes
- Spatial and Panel Data Analysis
- Statistical Methods in Clinical Trials
- Cardiac Imaging and Diagnostics
- Global Cancer Incidence and Screening
- Primary Care and Health Outcomes
- Emergency and Acute Care Studies
- Bayesian Modeling and Causal Inference
- Orthopaedic implants and arthroplasty
- Research in Social Sciences
University of Limerick
2009-2024
Binus University
2024
Ninewells Hospital
2024
Princess Margaret Hospital for Children
2024
Perth Children's Hospital
2024
École Nationale de la Statistique et de l'Administration Économique
2010-2023
École Nationale de la Statistique et de l'Analyse de l'Information
2010-2023
Centre de Recherche en Économie et Statistique
2016-2020
Iowa State University
2017
University of Kansas
2017
death') is now recognized as a clinical entity and existing knowledge has been recently sum- marized (Valdes-
Journal Article On modelling mean‐covariance structures in longitudinal studies Get access Jianxin Pan, Pan Search for other works by this author on: Oxford Academic Google Scholar Gilbert Mackenzie Biometrika, Volume 90, Issue 1, March 2003, Pages 239–244, https://doi.org/10.1093/biomet/90.1.239 Published: 01 2003
Two hundred and six patients with acute stroke admitted consecutively to District General Hospitals, were studied for a period of months. Significance tests conducted singly detected 21 factors present during the first 48 h stroke, which related outcome months later in terms both mortality functional recovery. Among these significant various measures perceptual dysfunction, including Albert's Test. Multivariate statistical analysis included discriminant linear logistic modelling, revealed...
One hundred and thirty seven previously untreated out-patients with first second degree haemorrhoids were allocated at random to treatment by infrared coagulation (n=66) or rubber band ligation (n=71). Complete follow up was obtained in 122 patients (60 who had undergone (group 1), 62 2)) periods from three months one year after completion of treatment. Infrared produced a satisfactory outcome 51 (85%): 34 rendered asymptomatic 17 improved. Rubber 57 (92%): 33 24 Both methods equally...
A preliminary report indicated stable or improved vision in 12 of 19 patients with subfoveal choroidal neovascularisation treated 15 Gy 6 MV photons to the affected macula after an average follow up 18 months. Here prolonged findings this group is reported which was further increased 41.Forty one were 10, 12, eye. Thirteen eyes also observed as a non-randomised comparison group.At 18, and 24 months mean change visual acuity radiotherapy less than 1 Bailey-Lovie line from that measured at...
Despite the use of standardized protocols in, multi-centre, randomized clinical trials, outcome may vary between centres. Such heterogeneity alter interpretation and reporting treatment effect. Below, we propose a general frailty modelling approach for investigating, inter alia, putative treatment-by-centre interactions in time-to-event data multi-centre trials. A correlated random effects model is used to baseline risk effect across It be based on shared, individual or effects. For...
Ten institutions in nine countries joined together to test the stability of resistance 14 potato genotypes oomycete pathogen Phytophthora infestans three separate trials. Seven were tested one trial involving seven locations, and all two subsequent trials, each eight locations. Stability was with nonparametric tests an additive main effects multiplicative interaction (AMMI) model. Overall, P. robust; resistant consistently locations The analysis indicated that specific basically stable...
Eighty-one patients with cancer of the male breast were studied. The majority (79) presented a mass in and 8 tumor was found by chance. Two serosanguinous discharge. Average duration symptoms 11.9 months. Thirty had Stage I, 25 II, 16 III, IV disease. Fifty-three simple mastectomy, nine lumpectomy, six radical five biopsy only. Eight no local surgery. Overall 5- 10-year survival allowing for all causes death 38% 17%, respectively. Cox's (1972) proportional hazard regression model used to...
Serum myoglobin concentration and creatine kinase activity were measured serially in 70 consecutive patients presenting within four hours of the onset symptoms suspected acute myocardial infarction. Of 36 with definite or possible infarction (WHO criteria), serum was raised (greater than 85 micrograms/l) one hour after 25% at 89%. Creatine 140 U/l) only 56%. Within 12 reached a peak 83% 14%. fell to normal values 67% while 3%. Four distinguished easily those from without, whereas when used...
In recent years external beam radiotherapy (EBRT) has been proposed as a treatment for the wet form of age-related macular degeneration (AMD) where choroidal neovascularization (CNV) is hallmark. While majority pilot (Phase I) studies have reported encouraging results, few found no benefit, i.e. EBRT was not to result in either improvement or stabilization visual acuity treated eye. The natural history loss untreated CNV AMD highly variable. Loss vision influenced mainly by presenting...
Abstract: Objective: To investigate the effect of haematinic status on erythropoietin throughout pregnancy in order to assess whether supplementation is appropriate for all pregnant women or should be reserved individuals whom deficiency has been identified. Methods: A prospective, repeated‐measures, longitudinal study 263 enrolled at an antenatal clinic a district general hospital who received standard obstetric management including supplements. Haematological indices, iron status, folate...
Various frailty models have been developed and are now widely used for analysing multivariate survival data. It is therefore important to develop an information criterion model selection. However, in there several alternative ways of forming a the particular chosen may not be uniformly best. In this paper, we study Akaike (AIC) on selecting structure from set (possibly) non-nested models. We propose two new AIC criteria, based conditional likelihood extended restricted (ERL) given by Lee...
A convenient reparametrization of the marginal covariance matrix arising in longitudinal studies is discussed. The new parameters have transparent statistical interpretations, are unconstrained and may be modelled parsimoniously terms polynomials time. We exploit this framework to model dependence structure on baseline covariates, time their interaction. rationale based assumption that a homogeneous with respect covariate space testable choice. Accordingly, we provide methods for testing by...
<b>Objective</b> To compare outcomes between adjustable spectacles and conventional methods for refraction in young people. <b>Design</b> Cross sectional study. <b>Setting</b> Rural southern China. <b>Participants</b> 648 people aged 12-18 (mean 14.9 (SD 0.98)), with uncorrected visual acuity ≤6/12 either eye. <b>Interventions</b> All participants underwent self without cycloplegia (paralysis of near focusing ability topical eye drops), automated cycloplegia, subjective by an ophthalmologist...
It is standard practice for covariates to enter a parametric model through single distributional parameter of interest, example, the scale in many survival models. Indeed, well-known proportional hazards this kind. In article, we discuss more general approach whereby than one simultaneously (e.g., and shape parameters). We refer as "multi-parameter regression" (MPR) modeling explore its use analysis context. find that multi-parameter regression leads flexible models which can offer greater...
It can be more challenging to efficiently model the covariance matrices for multivariate longitudinal data than univariate case, due correlations arising between multiple responses. The positive-definiteness constraint and high dimensionality are further obstacles in modelling. In this paper, we develop a data-based method by which parameters replaced unconstrained interpretable with reduced dimensions. maximum likelihood estimators mean shown consistent asymptotically normally distributed....