Portability of an algorithm to identify rheumatoid arthritis in electronic health records

By Robert J. Carroll, William K. Thompson, A. Eyler, A. Mandelin, Tianxi Cai, Raquel M. Zink, J. Pacheco, C. Boomershine, Thomas A. Lasko, Hua Xu and 10 more. 2012

230 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Complementary and alternative medicine, Rheumatology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Complementary and alternative medicine

David A. Sontag

In Rheumatology

Citations in the corpus

Natural Language Processing for EHR-Based Computational Phenotyping (2018, background): “[35] HITEx, Customized NegEx queries HITEx Diagnosis, billing, medication, procedural codes, physician text notes, discharge summaries, laboratory test results, radiology report 21 defined attributes from patients’ narrative”
Methods to Develop an Electronic Medical Record Phenotype Algorithm to Compare the Risk of Coronary Artery Disease across 3 Chronic Disease Cohorts (2015, background): “The RA phenotype algorithm has a PPV of 94% and contains 4,453 subjects [8, 12].”
Development of phenotype algorithms using electronic medical records and incorporating natural language processing (2015, methodology): “13 We thank the i2b2 team members integral to the development of our EMR algorithms, including Andrew Cagan, programmer, Research Computing, Partners Healthcare; Su-Chun Cheng, senior research scientist, Harvard School of Public Health; Sergey Goryachev, software developer, Ariadne Labs; Vishesh…”
Natural Language Processing of Clinical Notes on Chronic Diseases: Systematic Review (2019, unknown): “Three studies concern diseases of the musculoskeletal system and connective tissue, in particular classification of snippets of text related to axial spondyloarthritis in the EMRs of US military veterans using NLP and SVM [94], phenotyping systemic lupus erythematosus [95], and identification of rhe”
Automatic identification of methotrexate-induced liver toxicity in patients with rheumatoid arthritis from the electronic medical record (2015, background): “The algorithm combined variables from NLP and codified EMR data, achieved a high accuracy (AUC 93.7%, positive predictive value (PPV) 94%, and sensitivity 63%, when specificity was set at 97%), and was portable across three academic hospital EMRs.15 Lin et al11,21,22 further explored multiple featur”
Applying deep neural networks to unstructured text notes in electronic medical records for phenotyping youth depression (2017, methodology): “10 NLP methods have also been applied to EMRs to boost the efficiency of manual chart abstraction for breast cancer recurrence with 92% sensitivity and 96% sensitivity.”
Extracting research-quality phenotypes from electronic health records to support precision medicine (2015, background): “created, validated and implemented throughout the network, and the results of genetic replications have been published [36,56-58].”
Extracting information from the text of electronic medical records to improve case detection: a systematic review (2016, methodology): “Such algorithms take significant human effort and time to develop, requiring domain expertise, programming skills, and iterative evaluation and development.(42) Re-using existing algorithms and nomenclatures minimizes effort and ensures comparability with other studies.”
Autoantibodies, autoimmune risk alleles and clinical associations in rheumatoid arthritis cases and non-RA controls in the electronic medical records (2013, unknown): “This algorithm was further validated at 2 other academic institutions (16) (see ref.”
Electronic medical record phenotyping using the anchor and learn framework (2016, background): “In addition, the learned classifiers are institution-specific and often do not generalize well without modifications(20) or local retraining.(24) Methodologically, the closest system to our current work uses “silver standard training sets,”(25,26) with partially reliable labels, within a machine lea”
Genome- and Phenome-Wide Analysis of Cardiac Conduction Identifies Markers of Arrhythmia Risk (2013, background): “One of the key lessons in the eMERGE experience to date has been that algorithms to identify cases and controls for genomic or other study can be successfully deployed across multiple EMR systems.(16,42,43) Thus, as the number of subjects with dense genomic information across multiple EMR systems gr”
Comparing deep learning and concept extraction based methods for patient phenotyping from clinical narratives (2018, methodology): “[13] describe the process of extraction, rule-generation and prediction as the general approach to patient phenotyping using the cTAKES [14, 51–53], and test this approach on various data sets [54].”
PheKB: a catalog and workflow for creating electronic phenotype algorithms for transportability (2016, methodology): “22 23–26 The Phenotype Knowledgebase (PheKB; available at http://phekb.org) was created as a workflow management system and learning center supporting the creation, validation, and dissemination of computable algorithms.”