MCN: A comprehensive corpus for medical concept normalization

Luo, Yen-Fu; Sun, Weiyi; Rumshisky, Anna

doi:10.1016/j.jbi.2019.103132

Cited by 42 publications

(53 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are over 3.5 million concepts in UMLS, and for each concept, UMLS also provides the definition, preferred term, synonyms, semantic type, relationships with other concepts, etc. Following the procedure in Luo et al, 10 we restrict our concepts to the 2 vocabularies of SNOMED-CT and RxNorm in UMLS version 2017AB. In our experiments, we make use of UMLS preferred terms, synonyms, and semantic types of these concepts.…”

Section: Methodsmentioning

confidence: 99%

“…As one of the most comprehensive biomedical ontologies, the Unified Medical Language System (UMLS) 5 has been widely used in CN such as clinical disorder normalization in 2013 ShARe/CLEF eHealth 6 and 2014 SemEval Task 7 Analysis of Clinical Text, 7 and adverse drug reaction normalization in Social Media Mining for Health shared tasks. 8 , 9 The organizers of the 2019 n2c2 (National NLP Clinical Challenges) shared task track 3 Clinical Concept Normalization adopted the medical concept normalization (MCN) corpus, 10 which is annotated with a broader set of medical concepts than the disease or disorder concepts of previous work. The work described in the current article is based on our participation in the 2019 n2c2 shared task.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Unified Medical Language System resources improve sieve-based generation and Bidirectional Encoder Representations from Transformers (BERT)–based ranking for concept normalization

Gopale

Zhang

et al. 2020

Journal of the American Medical Informatics Association

View full text Add to dashboard Cite

Objective Concept normalization, the task of linking phrases in text to concepts in an ontology, is useful for many downstream tasks including relation extraction, information retrieval, etc. We present a generate-and-rank concept normalization system based on our participation in the 2019 National NLP Clinical Challenges Shared Task Track 3 Concept Normalization. Materials and Methods The shared task provided 13 609 concept mentions drawn from 100 discharge summaries. We first design a sieve-based system that uses Lucene indices over the training data, Unified Medical Language System (UMLS) preferred terms, and UMLS synonyms to generate a list of possible concepts for each mention. We then design a listwise classifier based on the BERT (Bidirectional Encoder Representations from Transformers) neural network to rank the candidate concepts, integrating UMLS semantic types through a regularizer. Results Our generate-and-rank system was third of 33 in the competition, outperforming the candidate generator alone (81.66% vs 79.44%) and the previous state of the art (76.35%). During postevaluation, the model’s accuracy was increased to 83.56% via improvements to how training data are generated from UMLS and incorporation of our UMLS semantic type regularizer. Discussion Analysis of the model shows that prioritizing UMLS preferred terms yields better performance, that the UMLS semantic type regularizer results in qualitatively better concept predictions, and that the model performs well even on concepts not seen during training. Conclusions Our generate-and-rank framework for UMLS concept normalization integrates key UMLS features like preferred terms and semantic types with a neural network–based ranking model to accurately link phrases in text to UMLS concepts.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Unified Medical Language System resources improve sieve-based generation and Bidirectional Encoder Representations from Transformers (BERT)–based ranking for concept normalization

Gopale

Zhang

et al. 2020

Journal of the American Medical Informatics Association

View full text Add to dashboard Cite

show abstract

“…The three most common information extraction tasks -named entity recognition (38,39,40), concept normalization (41,42), and relation extraction (Section 7) -are still active areas of research. However, in many cases, software systems exist that will perform these tasks automatically.…”

Section: Software For Clinical Information Extractionmentioning

confidence: 99%

Modern Clinical Text Mining: A Guide and Review

Percha¹

2020

Preprint

View full text Add to dashboard Cite

Electronic health records (EHRs) are becoming a vital source of data for healthcare quality improvement, research, and operations. However, much of the most valuable information contained in EHRs remains buried in unstructured text. The field of clinical text mining has advanced rapidly in recent years, transitioning from rule-based approaches to machine learning and, more recently, deep learning. With new methods come new challenges, however, especially for those new to the field. This review provides an overview of clinical text mining for those who are encountering it for the first time (e.g. physician researchers, operational analytics teams, machine learning scientists from other domains). While not a comprehensive survey, it describes the state of the art, with a particular focus on new tasks and methods developed over the past few years. It also identifies key barriers between these remarkable technical advances and the practical realities of implementation at health systems and in industry.

show abstract

“…It was first proposed for coreference resolution [30] and then applied to other tasks such as temporal relations and quote attribution [31,32]. It has also been employed for various tasks in the biomedical domains, such as entity linking, coreference resolution, relation extraction, and concept normalization [33][34][35][36]. Our system can be seen as consisting of four sieves that capture different types of information: (1) context triggers, (2) short-range intra-sentence (i.e., intraclause) events, (3) long-range intra-sentence (i.e., crossclause) events, and (4) cross-sentence events.…”

Section: Multi-pass Sieve Architecturementioning

confidence: 99%

“…This means that it is also easy to assess the contribution of each sieve to the overall performance. This is particularly useful when supervised learning does not work well, as discussed in the section on related work [30][31][32][33][34][35][36].…”

Section: Multi-pass Sieve Architecturementioning

confidence: 99%

Unsupervised inference of implicit biomedical events using context triggers

2020

View full text Add to dashboard Cite

Background: Event extraction from the biomedical literature is one of the most actively researched areas in biomedical text mining and natural language processing. However, most approaches have focused on events within single sentence boundaries, and have thus paid much less attention to events spanning multiple sentences. The Bacteria-Biotope event (BB-event) subtask presented in BioNLP Shared Task 2016 is one such example; a significant amount of relations between bacteria and biotope span more than one sentence, but existing systems have treated them as false negatives because labeled data is not sufficiently large enough to model a complex reasoning process using supervised learning frameworks. Results: We present an unsupervised method for inferring cross-sentence events by propagating intra-sentence information to adjacent sentences using context trigger expressions that strongly signal the implicit presence of entities of interest. Such expressions can be collected from a large amount of unlabeled plain text based on simple syntactic constraints, helping to overcome the limitation of relying only on a small number of training examples available. The experimental results demonstrate that our unsupervised system extracts cross-sentence events quite well and outperforms all the state-of-the-art supervised systems when combined with existing methods for intra-sentence event extraction. Moreover, our system is also found effective at detecting long-distance intra-sentence events, compared favorably with existing high-dimensional models such as deep neural networks, without any supervised learning techniques. Conclusions: Our linguistically motivated inference model is shown to be effective at detecting implicit events that have not been covered by previous work, without relying on training data or curated knowledge bases. Moreover, it also helps to boost the performance of existing systems by allowing them to detect additional cross-sentence events. We believe that the proposed model offers an effective way to infer implicit information beyond sentence boundaries, especially when human-annotated data is not sufficient enough to train a robust supervised system.

show abstract

MCN: A comprehensive corpus for medical concept normalization

Cited by 42 publications

References 11 publications

Unified Medical Language System resources improve sieve-based generation and Bidirectional Encoder Representations from Transformers (BERT)–based ranking for concept normalization

Unified Medical Language System resources improve sieve-based generation and Bidirectional Encoder Representations from Transformers (BERT)–based ranking for concept normalization

Modern Clinical Text Mining: A Guide and Review

Unsupervised inference of implicit biomedical events using context triggers

Contact Info

Product

Resources

About