A5041910623- Statistical Methods and Inference
- Statistical Methods and Bayesian Inference
- Bayesian Methods and Mixture Models
- Advanced Statistical Methods and Models
- Gaussian Processes and Bayesian Inference
- Bayesian Modeling and Causal Inference
- Advanced Statistical Process Monitoring
- Liver Disease Diagnosis and Treatment
- Dementia and Cognitive Impairment Research
- Orthopaedic implants and arthroplasty
- Cardiac, Anesthesia and Surgical Outcomes
- Stroke Rehabilitation and Recovery
- Total Knee Arthroplasty Outcomes
- Cardiac Arrest and Resuscitation
- Prostate Cancer Treatment and Research
- Data Management and Algorithms
- Traumatic Brain Injury Research
- Cardiac Structural Anomalies and Repair
- Cardiac Valve Diseases and Treatments
- Acute Myocardial Infarction Research
- Ophthalmology and Visual Health Research
- Cancer-related cognitive impairment studies
- Quality and Safety in Healthcare
- Topic Modeling
- Control Systems and Identification
Amsterdam University Medical Centers
2021-2025
Public Health Service of Amsterdam
2022-2025
Vrije Universiteit Amsterdam
2021-2025
Leiden University Medical Center
2017-2020
Leiden University
2014-2020
Loyola University Medical Center
2020
Mathematical Institute of the Slovak Academy of Sciences
2014-2019
Budapest University of Technology and Economics
2017
University of Amsterdam
2014
Cancer Research And Biostatistics
2013
We consider the horseshoe estimator due to Carvalho, Polson and Scott (2010) for multivariate normal mean model in situation that vector is sparse nearly black sense. assume frequentist framework where data generated according a fixed vector. show if number of nonzero parameters known, attains minimax $\ell_{2}$ risk, possibly up multiplicative constant. provide conditions under which combined with an empirical Bayes estimate means still yields risk. furthermore prove upper bound on rate...
We begin by introducing the main ideas of paper under discussion. discuss some interesting issues regarding adaptive component-wise credible intervals. then briefly touch upon concepts self-similarity and excessive bias restriction. This is followed comments on extensive simulation study carried out in paper.
We investigate the frequentist properties of Bayesian procedures for estimation based on horseshoe prior in sparse multivariate normal means model. Previous theoretical results assumed that sparsity level, is, number signals, was known. drop this assumption and characterize behavior maximum marginal likelihood estimator (MMLE) a key parameter prior. prove MMLE is an effective sense it leads to (near) minimax optimal underlying mean vector generating data. Besides empirical Bayes procedure,...
The first Bayesian results for the sparse normal means problem were proven spike-and-slab priors. However, these priors are less convenient from a computational point of view. In meanwhile, large number continuous shrinkage has been proposed. Many can be written as scale mixture normals, which makes them particularly easy to implement. We propose general conditions on prior local variance in mixtures such that posterior contraction at minimax rate is assured. require tails least heavy...
Since their inception in the 1980s, regression trees have been one of more widely used nonparametric prediction methods. Tree-structured methods yield a histogram reconstruction surface, where bins correspond to terminal nodes recursive partitioning. Trees are powerful, yet susceptible overfitting. Strategies against overfitting traditionally relied on pruning greedily grown trees. The Bayesian framework offers an alternative remedy through priors. Roughly speaking, good prior charges...
Abstract Combining data from various sources empowers researchers to explore innovative questions, for example those raised by conducting healthcare monitoring studies. However, the lack of a unique identifier often poses challenges. Record linkage procedures determine whether pairs observations collected on different occasions belong same individual using partially identifying variables (e.g. birth year, postal code). Existing methodologies typically involve compromise between computational...
Importance Hip fractures in older adults are serious injuries that result disability, higher rates of illness and death, a substantial strain on health care resources. High-quality evidence to improve hip fracture regarding the surgical approach hemiarthroplasty is lacking. Objective To compare 6-month outcomes posterolateral (PLA) direct lateral (DLA) for patients with acute femoral neck fracture. Design, Setting, Participants This multicenter, randomized clinical trial (RCT) comparing DLA...
Background and purpose - In arthroplasty registry studies, the analysis of time to revision is complicated by competing risk death. There are no clear guidelines for choice between 2 main adjusted methods, cause-specific Cox Fine-Gray regression, orthopedic data. We investigated whether there benefits, such as insight into different aspects progression revision, using either 1 or both regression methods in studies general, specifically when length follow-up short relative expected survival...
Osteogenesis Imperfecta (OI) is a complex disease caused by genetic alterations in production of collagen type I, and collagen-related proteins. Bone fragility the most common patient issue, but extraskeletal complications also present an adverse factor quality life prognosis patients with OI. However, still little known about morbidity mortality these patients. The objective this paper to determine describe what extent OI impacts patients’ terms hospitalization describing incidence...
Background: Colorectal anastomotic leakage (CAL) remains a severe complication after bowel surgery. Recent research has identified several modifiable perioperative CAL risk factors. The aim of this study is to implement an enhanced care protocol, optimize the intraoperative condition patient and minimalize exposure Secondly, will investigate whether implementation new protocol results in decreased rate. Methods: DOUBLE CHECK be performed using open-label, international, multicenter design 11...
We introduce a Bayesian estimator of the underlying class structure in stochastic block model, when number classes is known. The posterior mode corresponding to Dirichlet prior on proportions, generalized Bernoulli labels, and beta edge probabilities. show that this strongly consistent expected degree at least order log2n, where n nodes network.
Abstract Background During COVID-19 lockdown measures, memory clinic patients reported worries for faster cognitive decline, due to loss of structure and feelings loneliness depression. We aimed investigate the impact on rate decline in a mixed population, compared matched historical controls. Methods included who visited Alzheimer Center Amsterdam 6 months 1 week before first Dutch lockdown, had second visit year later, after this period ( n = 113; 66 ± 7 years old; 30% female; 55 dementia,...
Barlow's disease (BD) is characterized by thick, redundant mitral valve (MV) leaflets, which can lead to prolapse and significant regurgitation (MR). MV annular abnormalities are also commonly observed increasingly recognized as possible primary pathology, with leaflet thickening being secondary increased stress on the apparatus. To provide more insights into this hypothesis, evolution of over time in patients BD was assessed.A total 64 (54 ± 12 years, 72% male) who underwent surgery had...
Introduction The aim was to examine the association between physical and mental comorbidity with receiving low vision services (LVS). Methods A retrospective study based on Dutch claims data of health insurers performed. We retrieved (2015–2018) patients (≥18 years) eye diseases causing severe loss who received LVS at rehabilitation organizations in 2018 (target group) did not receive LVS, but ophthalmic medical specialist care for glaucoma, macular, diabetic retinal and/or (reference...
Background/Aims: Randomisation in small clinical trials is a delicate matter, due to the tension between conflicting aims of balanced groups and unpredictable allocations. The commonly used method permuted block randomisation has been heavily criticised for its high predictability. This article introduces merged randomisation, novel conceptually simple restricted design (less than 100 patients per stratum). Merged procedure that can be carried out without need computer. not 1:1 but readily...
To describe the course of depressive symptoms during first 12 months post-stroke and its association with unmet needs.A prospective cohort study among stroke patients admitted to inpatient rehabilitation. Depressive were assessed 3, 6, using Hospital Anxiety Depression Scale, categorized into three trajectories: no (all times <8), non-consistent (one or two ≥8), persistent ≥8) symptoms. Unmet needs Longer-Term Needs questionnaire. Multivariable logistic regression analyses used investigate...
We study the switch distribution, introduced by van Erven, Grünwald and De Rooij (2012), applied to model selection subsequent estimation.While switching was known be strongly consistent, here we show that it achieves minimax optimal parametric risk rates up a log n factor when comparing two nested exponential families, partially confirming conjecture Lauritzen (2012) Cavanaugh behaves asymptotically like Hannan-Quinn criterion.Moreover, Bayes selection, but unlike standard significance...
Summary Rapidly detecting problems in the quality of care is utmost importance for well-being patients. Without proper inspection schemes, such can go undetected years. Cumulative sum (CUSUM) charts have proven to be useful control, yet available methodology survival outcomes limited. The few continuous time usually require researcher specify an expected increase failure rate advance, thereby requiring prior knowledge about problem at hand. Misspecifying parameters lead false positive alerts...
In arthroplasty data, patients with staged bilateral total joint (TJA) pose a problem in statistical analysis. Subgroup analysis, which unilateral and TJA are studied separately, is sometimes considered an appropriate solution to the problem; we aim show that this not true because of immortal time bias.We reviewed who underwent (at any time) TJA. The logical fallacy leading bias explained through simple artificial data example. cumulative incidences revision death computed by subgroup...