Background and Aims Proton pump inhibitors (PPIs) have been associated with adverse clinical outcomes amongst clopidogrel users after an acute coronary syndrome. Recent pre-clinical results suggest that this risk might extend to subjects without any prior history of cardiovascular disease. We explore potential in the general population via data-mining approaches. Methods Using a novel approach for mining data pharmacovigilance, we queried over 16 million documents on 2.9 individuals examine...

Signal-detection algorithms (SDAs) are recognized as vital tools in pharmacovigilance. However, their performance characteristics generally unknown. By leveraging a unique gold standard recently made public by the Observational Medical Outcomes Partnership (OMOP) and conducting systematic evaluation, we provide new insights into diagnostic potential of SDAs that routinely applied to US Food Drug Administration (FDA) Adverse Event Reporting System (AERS). We find can attain reasonable...

Background— Proton pump inhibitors (PPIs) are gastric acid–suppressing agents widely prescribed for the treatment of gastroesophageal reflux disease. Recently, several studies in patients with acute coronary syndrome have raised concern that use PPIs these may increase their risk major adverse cardiovascular events. The mechanism this possible effect is not known. Whether general population might also be at has been addressed. Methods and Results— Plasma asymmetrical dimethylarginine (ADMA)...

With increasing adoption of electronic health records (EHRs), there is an opportunity to use the free-text portion EHRs for pharmacovigilance. We present novel methods that annotate unstructured clinical notes and transform them into a deidentified patient–feature matrix encoded using medical terminologies. demonstrate resulting high-throughput data detecting drug–adverse event associations adverse events associated with drug–drug interactions. show these flag early (in most cases before...

Electronic health records (EHRs) are increasingly being used to complement the FDA Adverse Event Reporting System (FAERS) and enable active pharmacovigilance. Over 30% of all adverse drug reactions caused by drug-drug interactions (DDIs) result in significant morbidity every year, making their early identification vital. We present an approach for identifying DDI signals directly from textual portion EHRs.We recognize mentions event concepts over 50 million clinical notes two sites create a...

Objective Depression is a prevalent disorder difficult to diagnose and treat. In particular, depressed patients exhibit largely unpredictable responses treatment. Toward the goal of personalizing treatment for depression, we develop evaluate computational models that use electronic health record (EHR) data predicting diagnosis severity response

Abstract Electronic health records (EHR) represent a rich and relatively untapped resource for characterizing the true nature of clinical practice quantifying degree inter-relatedness medical entities such as drugs, diseases, procedures devices. We provide unique set co-occurrence matrices, pairwise mentions 3 million terms mapped onto 1 concepts, calculated from raw text 20 notes spanning 19 years data. Co-frequencies were computed by means parallelized annotation, hashing, counting...

The electronic surveillance for adverse drug events is largely based upon the analysis of coded data from reporting systems. Yet, vast majority health lies embedded within free text clinical notes and not gathered into centralized repositories. With increasing access to large volumes medical data-in particular notes-it may be possible computationally encode test safety signals in an active manner.We describe application simple annotation tools on mining resulting annotations compute risk...

Peripheral arterial disease (PAD) is a growing problem with few available therapies. Cilostazol the only FDA-approved medication class I indication for intermittent claudication, but carries black box warning due to concerns increased cardiovascular mortality. To assess validity of this warning, we employed novel text-analytics pipeline quantify adverse events associated use in clinical setting, including patients congestive heart failure (CHF).We analyzed electronic medical records 1.8...

Event sequences, such as patients' medical histories or users' sequences of product reviews, trace how individuals progress over time. Identifying common patterns, progression stages, in event is a challenging task because not every individual follows the same evolutionary pattern, stages may have very different lengths, and at rates. In this paper, we develop model-based method for discovering general sequences. We generative model which each sequence belongs to class, from given class pass...

Mental illness is the leading cause of disability in USA, but boundaries between different mental illnesses are notoriously difficult to define. Electronic medical records (EMRs) have recently emerged as a powerful new source information for defining phenotypic signatures specific diseases. We investigated how EMR-based text mining and statistical analysis could elucidate three important neuropsychiatric illnesses-autism, bipolar disorder, schizophrenia.We analyzed over 7000 patients at two...

Off-label drug use, defined as use of a in manner that deviates from its approved by the drug's FDA label, is problematic because such uses have not been evaluated for safety and efficacy. Studies estimate 21% prescriptions are off-label, only 27% those evidence We describe data-mining approach systematically identifying off-label usages using features derived free text clinical notes extracted two databases on known usage (Medi-Span DrugBank). trained highly accurate predictive model...

Abstract Objective The trade-off between the speed and simplicity of dictionary-based term recognition richer linguistic information provided by more advanced natural language processing (NLP) is an area active discussion in clinical informatics. In this paper, we quantify among text systems that make different trade-offs understanding. We tested both types three research tasks: phase IV safety profiling a drug, learning adverse drug–drug interactions, used-to-treat relationships drugs...

In this paper, we show that representation and reasoning techniques used in traditional knowledge engineering the emerging Semantic Web can play an important role for heterogeneous database integration. Our OntoGrate architecture combines ontology-based schema representation, first order logic inference, some SQL wrappers to integrate two sample relational databases. We define inferential data integration as theoretical framework our approach. The performance evaluation query answering shows...

To realize the Semantic Web, it will be necessary to make existing database content available for emerging Web applications, such as web agents and services, which use ontologies formally define semantics of their data. Our research in design implementation an ontology-based system, OntoGrate, addresses critical challenging problem supporting human experts multiple domains interactively integrate information that is heterogenous both structure semantics. Databases, knowledge bases, World...

Juvenile idiopathic arthritis is the most common rheumatic disease in children. Chronic uveitis a and serious comorbid condition of juvenile arthritis, with insidious presentation potential to cause blindness. Knowledge clinical associations will improve risk stratification. Based on observation, we hypothesized that allergic conditions are associated chronic patients.This study retrospective cohort using Stanford's data warehouse containing from Lucile Packard Children's Hospital 2000-2011...

Data mining is a crucial tool for identifying risk signals of potential adverse drug reactions (ADRs). However, ADR currently limited to leveraging single data source at time. It widely believed that combining evidence from multiple sources will result in more accurate identification system. We present methodology based on empirical Bayes modeling combine mined ~5 million event reports collected by the FDA, and healthcare corresponding 46 patients' main two types information employed signal...

As researchers analyze huge amounts of data that are annotated by large biomedical ontologies, one the major challenges for mining and machine learning is to leverage both ontologies together in a systematic scalable way. In this paper, we address two interesting related problems data: i) how discover semantic associations with help formal ii) identify potential errors data. By representing using RDF hyper graphs, subsequently transforming graphs corresponding bipartite forms, provide...