The Unified Medical Language System (UMLS): integrating biomedical terminology

By Olivier Bodenreider. 2004

4,811 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Genetics, Molecular Biology, Knowledge Graphs & Representation, Medical & Health AI

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 Genetics

In Molecular Biology

In Knowledge Graphs & Representation

In Medical & Health AI

Citations in the corpus

COMET: Commonsense Transformers for Automatic Knowledge Graph Construction (2019, neutral): “Knowledge base construction Previous work has looked at constructing knowledge bases as relational schemas using expert knowledge (Lenat, 1995; Bodenreider, 2004; Miller, 1995), semistructured text extraction (Suchanek et al.”
Natural Language Processing for EHR-Based Computational Phenotyping (2018, methodology): “They retrieved 6,928 phrases corresponding to the 80 concepts from the Unified Medical Language System (UMLS) Metathesaurus MRCONSO table [77].”
Graph-Sparse LDA: A Topic Model with Structured Sparsity (2014, methodology): “For this we use a diagn osis hierarchy [11] to recover clinically relevant subtype s described by a small set of concepts.”
Self-Alignment Pretraining for Biomedical Entity Representations (2020, methodology): “In all experiments, S AP B ERT denotes further pre-training with our self-alignment method on UMLS.”
A systematic comparison of feature space effects on disease classifier performance for phenotype identification of five diseases (2015, methodology): “One approach for semantically-informed machine learning-based phenotype identification leverages the fixed vocabulary of the Unified Medical Language System's (Bodenreider, 2004; Lindberg, Humphreys, & McCray, 1993) (UMLS) concept unique identifiers (CUIs).”
deepDR: a network-based deep learning approach to in silico drug repositioning (2019, methodology): “Chemical name, generic name or commercial name of each drug were standardized by Medical Subject Headings (MeSH) and Unified Medical Language System (UMLS) vocabularies (Bodenreider, 2004) and further converted to DrugBank ID from the DrugBank database (v4.”
The druggable genome and support for target identification and validation in drug development (2016, background): “FDBdrug indications are tagged with Universal Medical Language System (UMLS) concept identifiers (CUIs) and could be mapped into MeSH and other ontologies within the UMLS meta-thesaurus (51, 74).”
Comparing deep learning and concept extraction based methods for patient phenotyping from clinical narratives (2018, neutral): “Finally, our unstructured data could be combined with structured input, such as lab results, data from UMLS, or ICD-codes.”
Large language models are few-shot clinical information extractors (2022, methodology): “Medication Extraction Baseline For normalization, all entities were linked to the UMLS via the default string overlap functionality of ScispaCy (Bo-denreider, 2004).”
CORD-19: The Covid-19 Open Research Dataset (2020, methodology): “(2020) releases a supplementary dataset over CORD-19 by augmenting text with entity mentions predicted from multiple techniques, including ScispaCy and weak supervision based on the NLM’s Unified Medical Language System (UMLS) Metathesaurus (Bodenreider, 2004).”
Lessons from Natural Language Inference in the Clinical Domain (2018, methodology): “We used Metamap (Aronson and Lang, 2010) – a tool to identify medical concepts from text and map them to standard terminologies in the Uniﬁed Medical Language System (UMLS) (Bodenreider, 2004).”
PREDICT: a method for inferring novel drug indications with application to personalized medicine (2011, neutral): “In order to handle variations in disease names between OMIM and drug indication sources, we mapped disease names to UMLS concepts (Bodenreider, 2004), serving as a standardized terminology.”
Neural Temporal Relation Extraction (2017, background): “UMLS (Bodenreider, 2004) is a comprehensive ontology of clinical terminology (somewhat analogous to WordNet (Miller, 1995)) that includes most clinical terms and thus can be used as a lookup resource for clinical vocabulary.”
Feature Robustness in Non-stationary Health Records: Caveats to Deployable Model Performance in Common Clinical Machine Learning Tasks (2019, methodology): “CUI Code Spanning In a manner similar to that of Gong et al. (2017), we use the human-readable descriptions of the original 181 raw ItemIDs to automatically aggregate features into groups associated with Concept Unique Identifiers (CUIs) from the Unified Medical Language System (UMLS) (Bodenreider, ”