Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Natural Language Processing”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 109 records · Page 6Linked to original sources
PubMedDate not supplied

Experiments in concept modeling for radiographic image reports.

OBJECTIVE: Development of methods for building concept models to support structured data entry and image retrieval in chest radiography. DESIGN: An organizing model for chest-radiographic reporting was built by analyzing manually a set of natural-language chest-radiograph reports. During model building, clinician-informaticians judged alternative conceptual structures according to four criteria: content of clinically relevant detail, provision for semantic constraints, provision for canonical forms, and simplicity. The organizing model was applied in representing three sample reports in their entirety. To explore the potential for automatic model discovery, the representation of one sample report was compared with the noun phrases derived from the same report by the CLARIT natural-language processing system. RESULTS: The organizing model for chest-radiographic reporting consists of 62 concept types and 17 relations, arranged in an inheritance network. The broadest types in the model include finding, anatomic locus, procedure, attribute, and status. Diagnoses are modeled as a subtype of finding. Representing three sample reports in their entirety added 79 narrower concept types. Some CLARIT noun phrases suggested valid associations among subtypes of finding, status, and anatomic locus. CONCLUSIONS: A manual modeling process utilizing explicitly stated criteria for making modeling decisions produced an organizing model that showed consistency in early testing. A combination of top-down and bottom-up modeling was required. Natural-language processing may inform model building, but algorithms that would replace manual modeling were not discovered. Further progress in modeling will require methods for objective model evaluation and tools for formalizing the model-building process.

Computer Simulation↗
PubMed2006

Automated identification of adverse events related to central venous catheters.

Methods for surveillance of adverse events (AEs) in clinical settings are limited by cost, technology, and appropriate data availability. In this study, two methods for semi-automated review of text records within the Veterans Administration database are utilized to identify AEs related to the placement of central venous catheters (CVCs): a Natural Language Processing program and a phrase-matching algorithm. A sample of manually reviewed records were then compared to the results of both methods to assess sensitivity and specificity. The phrase-matching algorithm was found to be a sensitive but relatively non-specific method, whereas a natural language processing system was significantly more specific but less sensitive. Positive predictive values for each method estimated the CVC-associated AE rate at this institution to be 6.4 and 6.2%, respectively. Using both methods together results in acceptable sensitivity and specificity (72.0 and 80.1%, respectively). All methods including manual chart review are limited by incomplete or inaccurate clinician documentation. A secondary finding was related to the completeness of administrative data (ICD-9 and CPT codes) used to identify intensive care unit patients in whom a CVC was placed. Administrative data identified less than 11% of patients who had a CVC placed. This suggests that other methods, including automated methods such as phrase matching, may be more sensitive than administrative data in identifying patients with devices. Considerable potential exists for the use of such methods for the identification of patients at risk, AE surveillance, and prevention of AEs through decision support technologies.

Artificial Intelligence↗
PubMedDate not supplied

Image content extraction: application to MR images of the brain.

A system for automatically extracting image content features was developed that combines registration to a labeled atlas with natural language processing of free-text radiology reports. The system was then tested with T1-weighted, spoiled gradient-echo magnetic resonance (MR) imaging studies of the brain performed in nine patients. The locations of 599 structures were visually assessed by an experienced radiologist and compared with the locations indicated by automated output. The in-plane accuracy of the contours was subjectively evaluated as either good, moderate, or poor. The criterion for classifying a structure as correctly located was that 90% or more of all the images containing the structure had to be correctly identified. For 98% of the structures, the images identified by the automated algorithm agreed with those identified by the radiologist, and in 83% of cases, image contours showed a good in-plane overlap. The results of this validation study demonstrate that this combination of registration and natural language processing is accurate in identifying relevant images from brain MR imaging studies. However, the range of applicability of this technique has yet to be determined by applying the technique to a large number of studies.

Algorithms↗
PubMed2025

The role of chromatin state in intron retention: A case study in leveraging large scale deep learning models.

Complex deep learning models trained on very large datasets have become key enabling tools for current research in natural language processing and computer vision. By providing pre-trained models that can be fine-tuned for specific applications, they enable researchers to create accurate models with minimal effort and computational resources. Large scale genomics deep learning models come in two flavors: the first are large language models of DNA sequences trained in a self-supervised fashion, similar to the corresponding natural language models; the second are supervised learning models that leverage large scale genomics datasets from ENCODE and other sources. We argue that these models are the equivalent of foundation models in natural language processing in their utility, as they encode within them chromatin state in its different aspects, providing useful representations that allow quick deployment of accurate models of gene regulation. We demonstrate this premise by leveraging the recently created Sei model to develop simple, interpretable models of intron retention, and demonstrate their advantage over models based on the DNA language model DNABERT-2. Our work also demonstrates the impact of chromatin state on the regulation of intron retention. Using representations learned by Sei, our model is able to discover the involvement of transcription factors and chromatin marks in regulating intron retention, providing better accuracy than a recently published custom model developed for this purpose.

Deep Learning↗
PubMed2003

Electronic interpretation of chest radiograph reports to detect central venous catheters.

OBJECTIVE: To evaluate whether a natural language processing system, SymText, was comparable to human interpretation of chest radiograph reports for identifying the mention of a central venous catheter (CVC), and whether use of SymText could detect patients who had a CVC. DESIGN: To identify patients who had a CVC, we performed two surveys of hospitalized patients. Then, we obtained available reports from 104 patients who had a CVC during one of two cross-sectional surveys (ie, case-patients) and 104 randomly selected patients who did not have a CVC (ie, control-patients). SETTING: A 600-bed public teaching hospital. RESULTS: Chest radiograph reports were available from 124 of the 208 participants. Compared with human interpretation, SymText had a sensitivity of 95.8% and a specificity of 98.7%. The use of SymText to identify case- and control-patients resulted in a sensitivity of 43% and a specificity of 98%. Successful application of SymText varied significantly by venous insertion site (eg, a sensitivity of 78% for subclavian and a sensitivity of 3.7% for femoral). Twenty-six percent of the case-patients had a femoral CVC. CONCLUSIONS: Compared with human interpretation, SymText performed well in interpreting whether a report mentioned a CVC. In patient populations with less frequent CVC placement in femoral veins, the sensitivity for CVC detection likely would be higher. Applying a natural language processing system to chest radiograph reports may be a useful adjunct to other data sources to automate detection of patients who had a CVC.

Catheterization, Central Venous↗
PubMedDate not supplied

SciFinder Scholar 2006: an empirical analysis of research topic query processing.

Topical search queries in SciFinder Scholar are processed through an extensive set of natural language processing algorithms that greatly enhance the relevance and comprehensiveness of the search results. Little detailed documentation on these algorithms has been published. However, a careful examination of the highlighted hit terms coupled with a comparison of results from small variations in query language reveal much additional, useful information about these algorithms. An understanding of how these algorithms work can lead to better search results and explain many unexpected results, including differing hit counts for singular versus plural query words and phrases.

Algorithms↗
PubMed1998

Automatic prediction of trauma registry procedure codes from emergency room dictations.

Current natural language processing techniques for recognition of concepts in the electronic medical record have been insufficient to allow their broad use for coding information automatically. We have undertaken a preliminary investigation into the use of machine learning methods to recognize procedure codes from emergency room dictations for a trauma registry. Our preliminary results indicate moderate success, and we believe future enhancements with additional learning techniques and selected natural language processing approaches will be fruitful.

Artificial Intelligence↗
PubMed2003

Towards a medical question-answering system: a feasibility study.

Question-answering (QA) systems, as have been presented and evaluated in several TREC conferences, are the next generation of search engines. They combine 'traditional' Information Retrieval (IR) with Natural Language Processing (NLP) and Knowledge Engineering techniques to provide shorter, more precise answers to natural language questions. We study here the feasibility of such a system for French in the health care domain. In this purpose, we collected a corpus of student questions in oral surgery. We examined two enabling conditions: on the IR side, how to select the right keywords in a question to identify relevant material on the Web for answering this question, a prerequisite for success; and on the NLP side, whether the contents of the questions fit the conceptual model of an existing QA prototype, a favorable condition for rapid implementation. A manual Web search enabled us to devise automatable principles for building IR queries for these questions. Besides, we could design a semantic model, using UMLS Semantic Network relations, which is consistent with our prototype and covers 90% of the questions. However, the high specialization of the domain and the clinical orientation of the questions, joined with the more limited resources online in the French language, may restrain the quantity of Web material available for answering these questions.

Expert Systems↗
PubMed1999

A technique for semantic classification of unknown words using UMLS resources.

Natural Language Processing (NLP) is a tool for transforming natural text into codable form. Success of NLP systems is contingent on a well constructed semantic lexicon. However, creation and maintenance of these lexicons is difficult, costly and time consuming. The UMLS contains semantic and syntactic information of medical terms, which may be used to automate some of this task. Using UMLS resources we have observed that it is possible to define one semantic type by its syntactic combinations with other types in a corpus of discharge summaries. These patterns of combination can then be used to classify words which are not in the lexicon. The technique was applied to a corpus for a single semantic type and generated a list of 875 words which matched the classification criteria for that type. The words were ranked by number of patterns matched and the top 95 words were correctly typed with 80% accuracy.

Humans↗
PubMed2002

Two biomedical sublanguages: a description based on the theories of Zellig Harris.

Natural language processing (NLP) systems have been developed to provide access to the tremendous body of data and knowledge that is available in the biomedical domain in the form of natural language text. These NLP systems are valuable because they can encode and amass the information in the text so that it can be used by other automated processes to improve patient care and our understanding of disease processes and treatments. Zellig Harris proposed a theory of sublanguage that laid the foundation for natural language processing in specialized domains. He hypothesized that the informational content and structure form a specialized language that can be delineated in the form of a sublanguage grammar. The grammar can then be used by a language processor to capture and encode the salient information and relations in text. In this paper, we briefly summarize his language and sublanguage theories. In addition, we summarize our prior research, which is associated with the sublanguage grammars we developed for two different biomedical domains. These grammars illustrate how Harris' theories provide a basis for the development of language processing systems in the biomedical domain. The two domains and their associated sublanguages discussed are: the clinical domain, where the text consists of patient reports, and the biomolecular domain, where the text consists of complete journal articles.

Abstracting and Indexing↗
PubMed1997

Corpus-based identification and refinement of semantic classes.

Medical Language Processing (MLP), especially in specific domains, requires fine-grained semantic lexica. We examine whether robust natural language processing tools used on a representative corpus of a domain help in building and refining a semantic categorization. We test this hypothesis with ZELLIG, a corpus analysis tool. The first clusters we obtain are consistent with a model of the domain, as found in the SNOMED nomenclature. They correspond to coarse-grained semantic categories, but isolate as well lexical idiosyncrasies belonging to the clinical sub-language. Moreover, they help categorize additional words.

Classification↗
PubMed2002

The lexical properties of the gene ontology.

The Gene Ontology (GO) is a construct developed for the purpose of annotating molecular information about genes and their products. The ontology is a shared resource developed by the GO Consortium, a group of scientists who work on a variety of model organisms. In this paper we investigate the nature of the strings found in the Gene Ontology and evaluate them for their usefulness in natural language processing (NLP). We extend previous work that identified a set of properties that reliably identifies natural language phrases in the Unified Medical Language System (UMLS). The results indicate that a large percentage (79%) of GO terms are potentially useful for NLP applications. Some 35% of the GO terms were found in a corpus derived from the MEDLINE bibliographic database, and 27% of the terms were found in the current edition of the UMLS.

Genes↗
PubMed1995

Dutch medical language processing: discussion of a prototype.

This paper discusses a medical language processor for Dutch (DMLP). The DMLP forms part of the results of a joint project sponsored by the European Commission, which aims at the intelligent extraction and processing of information of patient discharge summaries. We will focus more specifically on the natural language processing aspects for Dutch which are treated by the K.U. Leuven team.

Language↗
PubMed1996

Variation in the application of natural processes: language-dependent constraints in the phonological acquisition of bilingual children.

This paper studies phonological processes and constraints on early phonological and lexical development, as well as the strategies employed by a young Spanish-, Portuguese-, and Hebrew-speaking child-Nurit (the author's niece)-in the construction of her early lexicon. Nurit's linguistic development is compared to that of another Spanish-, Portuguese-, and Hebrew-speaking child-Noam (the author's son). Noam and Nurit's linguistic development is contrasted to that of Berman's (1977) English- and Hebrew-speaking daughter (Shelli). The simultaneous acquisition of similar (closely related languages) such as Spanish and Portuguese versus that of nonrelated languages such as English and Hebrew yields different results: Children acquiring similar languages seem to prefer maintenance as a strategy for the construction of their early lexicon, while children exposed to nonrelated languages appear to prefer reduction to a large extent (Faingold, 1990). The Spanish- and Portuguese-speaking children's high accuracy stems from a wider choice of target words, where the diachronic development of two closely related languages provides a simplified model lexicon to the child.

Child Language↗
PubMed1995

Associating semantic grammars with the SNOMED: processing medical language and representing clinical facts into a language-independent frame.

We describe one approach for natural language processing of medical texts that associates a semantic grammar with the SNOMED (Systematized Nomenclature of Medicine). Our research hypothesis is that the combination of the nomenclature's declarative knowledge with a formal grammar would create a scientific sublanguage embedded with medical knowledge that could be used for analyzing and formatting medical texts. This combination permitted the abstraction of templates we call "semantic patterns." These patterns represent both linguistic and medical knowledge, packed into a hybrid information format. We analyzed manually case reports described in the New England Journal of Medicine (NEJM) from 1985 to 1988 and extracted empirically a semantic grammar. Over 2,000 sentences were analyzed. About 160 structural semantic patterns were abstracted and included in the database of one parser. We tested the parser using reports from 1989 to 1990. Results show that this approach is efficient for processing, indexing, and structuring diverse parts of case reports narrative. The analyzed medical sentences are structured into a language-independent semantic frame format. We conclude that the association of semantic grammars with the SNOMED enabled the construction of a formal system for analysis and representation of clinical facts. The transformation of the structured information from its frame format into other representational schemes, like conceptual graphs, is straightforward. Another application includes the use of the formatted language-independent frame for telegraphic English-Japanese translations of medical sentences.

Language↗
PubMed1993

UMLS knowledge for biomedical language processing.

This paper describes efforts to provide access to the free text in biomedical databases. The focus of the effort is the development of SPECIALIST, an experimental natural language processing system for the biomedical domain. The system includes a broad coverage parser supported by a large lexicon, modules that provide access to the extensive Unified Medical Language System (UMLS) Knowledge Sources, and a retrieval module that permits experiments in information retrieval. The UMLS Metathesaurus and Semantic Network provide a rich source of biomedical concepts and their interrelationships. Investigations have been conducted to determine the type of information required to effect a map between the language of queries and the language of relevant documents. Mappings are never straightforward and often involve multiple inferences.

Humans↗
PubMed2003

Integrating a hypernymic proposition interpreter into a semantic processor for biomedical texts.

Semantic processing provides the potential for producing high quality results in natural language processing (NLP) applications in the biomedical domain. In this paper, we address a specific semantic phenomenon, the hypernymic proposition, and concentrate on integrating the interpretation of such predications into a more general semantic processor in order to improve overall accuracy. A preliminary evaluation assesses the contribution of hypernymic propositions in providing more specific semantic predications and thus improving effectiveness in retrieving treatment propositions in MEDLINE abstracts. Finally, we discuss the generalization of this methodology to additional semantic propositions as well as other types of biomedical texts.

Abstracting and Indexing↗