A5103222064- Biomedical Text Mining and Ontologies
- Semantic Web and Ontologies
- Machine Learning and Data Classification
- Machine Learning in Healthcare
- Genomics and Rare Diseases
- Cardiac Valve Diseases and Treatments
- Explainable Artificial Intelligence (XAI)
- Topic Modeling
- Scientific Computing and Data Management
- Renal Diseases and Glomerulopathies
- Genetic Syndromes and Imprinting
- Natural Language Processing Techniques
- Cardiac Imaging and Diagnostics
- Genetic Associations and Epidemiology
- Connective tissue disorders research
- Atrial Fibrillation Management and Outcomes
- Dermatological and Skeletal Disorders
- Electronic Health Records Systems
- Research Data Management Practices
- Chronic Kidney Disease and Diabetes
- Acute Myocardial Infarction Research
- RNA Research and Splicing
- Recommender Systems and Techniques
- Lipoproteins and Cardiovascular Health
- Bioinformatics and Genomic Networks
University of Pennsylvania
2016-2021
Translational Therapeutics (United States)
2016-2020
University of Pennsylvania Health System
2016
Background: Truncating variants in the Titin gene (TTNtvs) are common individuals with idiopathic dilated cardiomyopathy (DCM). However, a comprehensive genomics-first evaluation of impact TTNtvs different clinical contexts, and modifiers such as genetic ancestry, has not been performed. Methods: We reviewed whole exome sequence data for >71 000 (61 040 from Geisinger MyCode Community Health Initiative (2007 to present) 10 273 PennMedicine BioBank (2013 identify anyone TTNtvs. further...
Biobanking necessitates extensive integration of data to allow analysis and specimen sharing. Ontologies have been demonstrated be a promising approach in fostering better semantic biobank-related data. Hitherto no ontology provided the coverage needed capture broad spectrum biobank user scenarios.Based principles laid out by Open Biological Biomedical Foundry two biobanking ontologies developed. These were merged using modular consistent with initial development principles. The merging was...
Background: Coronary artery disease (CAD) is influenced by genetic variation and traditional risk factors. Polygenic scores (PRS), which can be ascertained before the development of factors, have been shown to identify individuals at elevated CAD. Here, we demonstrate that a genome-wide PRS for CAD predicts all-cause mortality after accounting not only cardiovascular factors but also angiographic itself. Methods: Individuals who underwent coronary angiography were enrolled in an...
Many researchers with domain expertise are unable to easily apply machine learning (ML) their bioinformatics data due a lack of ML and/or coding expertise. Methods that have been proposed thus far automate mostly require programming experience as well expert knowledge tune and the algorithms correctly. Here, we study method automating biomedical science using web-based AI platform recommend model choices conduct experiments. We two goals in mind: first, make it easy construct sophisticated...
Standardizing clinical information in a semantically rich data model is useful for promoting interoperability and facilitating high quality research. Semantic Web technologies such as Resource Description Framework can be utilized to their full potential when accurately reflects the semantics of situation it describes. To this end, ontologies that abide by sound organizational principles used building blocks storage data. However, challenge programmatically define load from disparate...
<i>To the Editor:</i>I believe that Dr. Frank McGuckin may be in error suggesting proposal for standardization of nomenclature surgery chronic ear infection by an ad hoc committee 15 introduces semantic difficulty "because there is no definition what meant infection." The matter "chronic suppurative otitis media" was considered Committee on Conservation Hearing American Academy Ophthalmology and Otolaryngology. It unnecessary length would added to communication if this were included it...
Clinical research studies often leverage various heterogeneous data sources including patient electronic health record, online survey, and genomic data. We introduce a graph-based, integration query tool called Carnival. demonstrate its powerful ability to unify from these disparate create datasets for two studies: prevalence incidence case/control matches coronary artery disease controls Marfan syndrome. conclude with future directions Carnival development.
Abstract Standardizing clinical information in a common data model is important for promoting interoperability and facilitating high quality research. Semantic Web technologies such as Resource Description Framework can be utilized to their full potential when accurately reflects the reality of situation it describes. To this end, Open Biomedical Ontologies Foundry provides set ontologies that conform principles realism used create realism-based model. However, challenge programmatically...
Abstract Although NLP has been used to support cancer research more broadly, the development of algorithms extract evidence progression from clinical notes lung is still in its infancy. In this study, we trained supervised machine learning classifiers using rich semantic features detect and classify statements status radiology exams. Our classifier achieves high F1-scores for detecting discerning (0.80), stable (0.82), not relevant (0.92) sentences, demonstrating promising performance. We...