A5079036939- Biomedical Text Mining and Ontologies
- Semantic Web and Ontologies
- Computational Drug Discovery Methods
- Machine Learning in Healthcare
- Time Series Analysis and Forecasting
- Tryptophan and brain disorders
- Pharmacovigilance and Adverse Drug Reactions
- Pharmacogenetics and Drug Metabolism
- Natural Language Processing Techniques
- Mental Health Research Topics
- Artificial Intelligence in Healthcare
- Genetic Associations and Epidemiology
- Advanced Text Analysis Techniques
- Electronic Health Records Systems
- Drug Transport and Resistance Mechanisms
- Health Systems, Economic Evaluations, Quality of Life
- Genomics and Rare Diseases
- Schizophrenia research and treatment
- Medical Coding and Health Information
- Treatment of Major Depression
- Chronic Disease Management Strategies
- Topic Modeling
- Acne and Rosacea Treatments and Effects
- Cardiac, Anesthesia and Surgical Outcomes
- Psoriasis: Treatment and Pathogenesis
European Bioinformatics Institute
2021-2025
Addenbrooke's Hospital
2021-2022
Cambridge University Hospitals NHS Foundation Trust
2021-2022
European Molecular Biology Laboratory
2022
King's College London
2016-2019
Universidad Carlos III de Madrid
2013-2015
Abstract As populations get older and medicine consumption rises, the rate of concurrent drug use polypharmacy among patients is increasing. Polypharmacy known to complicate therapy increase risk drug-drug interactions, individuality which remain largely unexplored. Here, we perform a series genome-wide association studies identify variants associated with dosage changes during episodes therapy. We extracted in-hospital prescription records from 847,537 in population-wide Danish hospital...
The early detection of drug-drug interactions (DDIs) is limited by the diffuse spread DDI information in heterogeneous sources. Computational methods promise to play a key role identification and explanation DDIs on large scale. However, such rely availability computable representations describing relevant domain knowledge. Current modeling efforts have focused partial shallow domain, failing adequately support computational inference discovery applications. In this paper, we describe...
Despite the significant health impacts of adverse events associated with drug-drug interactions, no standard models exist for managing and sharing evidence describing potential interactions between medications. Minimal information have been used in other communities to establish community consensus around simple capable communicating useful information. This paper reports on a new minimal model interactions. A task force Semantic Web Health Care Life Sciences Community Group World-Wide...
Background: Anticholinergic medications are associated with adverse outcomes in older adults and should be prescribed cautiously. We describe the Risk Scale (ARS) scores of inpatients associations outcomes. Methods: included all emergency, first admissions ⩾65 years old admitted to one hospital over 4 years. Demographics, discharge specialty, dementia/history cognitive concern, illness acuity were retrieved from electronic records. ARS calculated as sum anticholinergic potential for each...
Electronic Health Records (EHR) data can provide novel insights into inpatient trajectories. Blood tests and vital signs from de-identified patients' hospital admission episodes (AE) were represented as multivariate time-series (MVTS) to train unsupervised Hidden Markov Models (HMM) represent each AE day one of 17 states. All HMM states clinically interpreted based on their patterns MVTS variables relationships with clinical information. Visualization differentiated patients progressing...
Abstract Electronic health record (EHR) systems with prescription data offer vast potential in pharmacoepidemiology and pharmacogenomics. The large amount of clinical recorded these requires automatic processing to extract relevant information. This paper introduces PRESNER, a name entity recognition (NER) classification pipeline for EHR data. uses the pre-trained transformer Bio-ClinicalBERT fine-tuned on UK Biobank entries manually annotated medication-related information (drug name, route...
Background: Animal studies suggest that the antibiotic and microglial activation inhibitor, minocycline, is likely to have a protective effect against emergence of psychosis but evidence from human lacking. The aim this study examine effects exposure minocycline during adolescence on later incidence severe mental illness (SMI). Methods: A historical cohort using electronic primary care data was conducted assess association between SMI. Incidence Rate Ratio (IRR) measured Poisson regression...
Natural Language Processing (NLP) techniques can provide an interesting way to mine the growing biomedical literature, and a promising approach for new knowledge discovery. However, major bottleneck in this area is that these systems rely on specific resources providing domain knowledge. Domain ontologies contextual framework semantic representation of domain, they contribute better performance current NLP systems. their contribution information extraction has not been well studied yet. The...
Natural language processing of pharmacological texts includes recognition drug names and extraction relationships between them. To this purpose, annotated corpora are required. These usually semantically by domain experts. However, other linguistic aspects should be considered to ensure the quality consistency annotation. This paper introduces several phenomena affecting annotation both named entities drug-drug interaction that arose during process DDI corpus. The detailed documentation...
Abstract Major depressive disorder is a complex condition with diverse presentations and polygenic underpinnings. Leveraging large biobanks linked to primary care prescription data, we developed data-driven approach based on antidepressant trajectories for patient stratification novel phenotype identification. We extracted quantitative 56,951 UK Biobank (UKB) 64,609 Danish National (CHB+DBDS) individuals. Using Hidden Markov Models K-means clustering, identified five six clusters,...
Background Minocycline has neurological anti-inflammatory properties and been hypothesised to have antipsychotic effects. Aim The aim of this study was investigate, using routinely collected United Kingdom primary health care data, whether adolescent men women are more or less likely receive an urgent psychiatric referral during treatment for acne with minocycline compared periods non-treatment. Method A self-controlled case series Clinical Practice Research Datalink calculate the incidence...
Abstract The implementation of Electronic Health Records (EHR) in UK hospitals provides new opportunities for clinical ‘big data’ analysis. representation observations routinely recorded practice is the first step to use these data several research tasks. Anonymised were extracted from 11 158 emergency admission episodes (AE) older adults. Irregular records 23 laboratory blood tests and vital signs normalized regularised into daily bins represented as numerical multivariate time-series...
Electronic Health Records (EHR) data can provide novel insights into inpatient trajectories. Blood tests and vital signs from de-identified patients’ hospital admission episodes (AE) were represented as multivariate time-series (MVTS) to train unsupervised Hidden Markov Models (HMM) represent each AE day one of 17 states. All HMM states clinically interpreted based on their patterns MVTS variables relationships with clinical information. Visualisation differentiated patients progressing...