A5040836504- Machine Learning in Healthcare
- Acute Ischemic Stroke Management
- Cerebrovascular and Carotid Artery Diseases
- Nutrition and Health in Aging
- Chronic Disease Management Strategies
- Health Systems, Economic Evaluations, Quality of Life
- COVID-19 Clinical Research Studies
- Rheumatoid Arthritis Research and Therapies
- Atrial Fibrillation Management and Outcomes
- Chronic Obstructive Pulmonary Disease (COPD) Research
- COVID-19 diagnosis using AI
- Cardiac, Anesthesia and Surgical Outcomes
- Colorectal Cancer Screening and Detection
- Total Knee Arthroplasty Outcomes
- Artificial Intelligence in Healthcare and Education
- Orthopaedic implants and arthroplasty
- Muscle Physiology and Disorders
- Exercise and Physiological Responses
- Quantum Information and Cryptography
- Quantum Computing Algorithms and Architecture
- Biosimilars and Bioanalytical Methods
- Lymphoma Diagnosis and Treatment
- COVID-19 and healthcare impacts
- Advanced Causal Inference Techniques
- Meta-analysis and systematic reviews
Erasmus MC
2019-2025
Deakin University
2023
Zealand University Hospital
2023
Erasmus University Rotterdam
2020-2022
Columbia University
2020
The Health-Analytics Data to Evidence Suite (HADES) is an open-source software collection developed by Observational Health Sciences and Informatics (OHDSI). It executes directly against healthcare data such as electronic health records administrative claims, that have been converted the Medical Outcomes Partnership (OMOP) Common Model. Using advanced analytics, HADES performs characterization, population-level causal effect estimation, patient-level prediction, potentially across a...
Abstract Background We investigated whether we could use influenza data to develop prediction models for COVID-19 increase the speed at which can reliably be developed and validated early in a pandemic. Estimated Risk (COVER) scores that quantify patient’s risk of hospital admission with pneumonia (COVER-H), hospitalization requiring intensive services or death (COVER-I), fatality (COVER-F) 30-days following diagnosis using historical from patients flu-like symptoms tested this patients....
Increasing evidence suggests that sarcopenia and a higher systemic immune-inflammation index (SII) are linked with morbidity in patients COPD. However, whether these two conditions contribute to all-cause mortality middle-aged older COPD or asthma is unclear. Therefore, we investigated the association between sarcopenia, SII, large-scale population-based setting.Between 2009 2014, 4482 participants (aged >55 years; 57.3% female) from Rotterdam Study were included. diagnosed clinically based...
Prognostic models help aid medical decision-making. Various prognostic are available via websites such as MDCalc, but these typically predict one outcome, for example, stroke risk. Each model requires individual predictors, age, lab results and comorbidities. There is no clinical tool to multiple outcomes from a list of common predictors. Identify constrained set outcome-agnostic We proposed novel technique aggregating the standardised mean difference across hundreds learn predictors that...
Hemorrhagic transformation (HT) after cerebral infarction is a complex and multifactorial phenomenon in the acute stage of ischemic stroke, often results poor prognosis. Thus, identifying risk factors making an early prediction HT contributes not only to selections therapeutic regimen but also, more importantly, improvement prognosis infarction. The purpose this study was develop validate model predict patient's within 30 days initial stroke.We utilized retrospective multicenter...
Abstract Background To demonstrate how the Observational Healthcare Data Science and Informatics (OHDSI) collaborative network standardization can be utilized to scale-up external validation of patient-level prediction models by enabling across a large number heterogeneous observational healthcare datasets. Methods Five previously published prognostic (ATRIA, CHADS 2 , VASC, Q-Stroke Framingham) that predict future risk stroke in patients with atrial fibrillation were replicated using OHDSI...
Objective To develop and externally validate COVID-19 Estimated Risk (COVER) scores that quantify a patient’s risk of hospital admission (COVER-H), requiring intensive services (COVER-I), or fatality (COVER-F) in the 30-days following diagnosis. Methods We analyzed federated network electronic medical records administrative claims data from 14 sources 6 countries. developed validated 3 using 6,869,127 patients with general practice, emergency room, outpatient visit diagnosed influenza...
Abstract Introduction Angiotensin converting enzyme inhibitors (ACEs) and angiotensin receptor blockers (ARBs) could influence infection risk of coronavirus disease (COVID-19). Observational studies to date lack pre-specification, transparency, rigorous ascertainment adjustment international generalizability, with contradictory results. Methods Using electronic health records from Spain (SIDIAP) the United States (Columbia University Irving Medical Center Department Veterans Affairs), we...
Identification of rheumatoid arthritis (RA) patients at high risk adverse health outcomes remains a major challenge. We aimed to develop and validate prediction models for variety in RA initiating first-line methotrexate (MTX) monotherapy. Data from 15 claims electronic record databases across 9 countries were used. Models developed internally validated on Optum® De-identified Clinformatics® Mart Database using L1-regularized logistic regression estimate the within 3 months (leukopenia,...
Abstract Background Prognostic models that are accurate could help aid medical decision making. Large observational databases often contain temporal data for large and diverse populations of patients. It may be possible to learn prognostic using the data. Often performance a model undesirably worsens when transported different database (or into clinical setting). In this study we investigate ensemble approaches combine independently developed (a simple federated learning approach) determine...
This study evaluates regularization variants in logistic regression (L1, L2, ElasticNet, Adaptive L1, Broken adaptive ridge [BAR], and Iterative hard thresholding [IHT]) for discrimination calibration performance, focusing on both internal external validation.
Background SARS-CoV-2 is straining health care systems globally. The burden on hospitals during the pandemic could be reduced by implementing prediction models that can discriminate patients who require hospitalization from those do not. COVID-19 vulnerability (C-19) index, a model predicts which will admitted to hospital for treatment of pneumonia or proxies, has been developed and proposed as valuable tool decision-making pandemic. However, at high risk bias according “prediction...
The purpose of this study was to develop and validate a prediction model for 90-day mortality following total knee replacement (TKR). TKR is safe cost-effective surgical procedure treating severe osteoarthritis (OA). Although complications surgery are rare, tools could help identify high-risk patients who be targeted with preventative interventions. aim simple inform treatment choices.A OA developed externally validated using US claims database UK general practice database. target population...
The standard approach to encoding constraints in quantum optimization is the quadratic penalty method. Quadratic penalties introduce additional couplings and energy scales, which can be detrimental performance of a optimizer. In annealing experiments performed on D-Wave Advantage, we explore an alternative method that only involves linear Ising terms apply it customer data science problem. Our findings support our hypothesis should improve compared using due its more efficient use physical...
Abstract The standard approach to encoding constraints in quantum optimization is the quadratic penalty method. Quadratic penalties introduce additional couplings and energy scales, which can be detrimental performance of a optimizer. In annealing experiments performed on D-Wave Advantage, we explore an alternative method that only involves linear Ising terms apply it customer data science problem. Our findings support our hypothesis should improve compared using due its more efficient use...
Sarcopenia is a heterogeneous skeletal muscle disorder involving the loss of mass and function. However, prevalence sarcopenia based on most recent definition remains to be determined in older people with chronic airway diseases. The aim was evaluate association diseases its lung function an population, using European Working Group Older People 2 (EWGSOP2) criteria. We performed cross-sectional analysis 5082 participants (mean age 69.0±8.8 years, 56% females) from Rotterdam Study....
External validation of prediction models is increasingly being seen as a minimum requirement for acceptance in clinical practice. However, the lack interoperability healthcare databases has been biggest barrier to this occurring on large scale. Recent improvements database enable standardized analytical framework model development and external validation. new lacks context, whereby can be compared with benchmark database. Iterative pairwise (IPEV) that uses rotating approach contextualize...
Abstract Background SARS-CoV-2 is straining healthcare systems globally. The burden on hospitals during the pandemic could be reduced by implementing prediction models that can discriminate between patients requiring hospitalization and those who do not. COVID-19 vulnerability (C-19) index, a model predicts which will admitted to hospital for treatment of pneumonia or proxies, has been developed proposed as valuable tool decision making pandemic. However, at high risk bias according...
We investigated a stacking ensemble method that combines multiple base learners within database. The results on external validation across four large databases suggest could improve model transportability.