PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Natural Language Processing”

Explore indexed PubMed citations for clinical trials, systematic reviews and public health research. Read source abstracts and follow each citation to its original PubMed record.

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

At least 73 records · Page 4Linked to original sources
PubMed1991

An automatic indexing method for medical documents.

This paper describes MetaIndex, an automatic indexing program that creates symbolic representations of documents for the purpose of document retrieval. MetaIndex uses a simple transition network parser to recognize a language that is derived from the set of main concepts in the Unified Medical Language System Metathesaurus (Meta-1). MetaIndex uses a hierarchy of medical concepts, also derived from Meta-1, to represent the content of documents. The goal of this approach is to improve document retrieval performance by better representation of documents. An evaluation method is described, and the performance of MetaIndex on the task of indexing the Slice of Life medical image collection is reported.

Abstracting and Indexing↗
PubMed1991

Latent Semantic Indexing of medical diagnoses using UMLS semantic structures.

The relational files within the UMLS Metathesaurus contain rich semantic associations to main concepts. We invoked the technique of Latent Semantic Indexing to generate information matrices based on these relationships and created "semantic vectors" using singular value decomposition. Evaluations were made on the complete set and subsets of Metathesaurus main concepts with the semantic type "Disease or Syndrome." Real number matrices were created with main concepts, lexical variants, synonyms, and associated expressions. Ancestors, children, siblings, and related terms were added to alternative matrices, preserving the hierarchical direction of the relation as the imaginary component of a complex number. Preliminary evaluation suggests that this technique is robust. A major advantage is the exploitation of semantic features which derive from a statistical decomposition of UMLS structures, possibly reducing dependence on the tedious construction of semantic frames by humans.

Abstracting and Indexing↗
PubMed1991

Establishing criterion validity of a computer-based clinical simulation.

Clinical simulations, depictions of real-life patient care situations, have been used extensively in medical education. This study determined the criterion validity of a computer-based simulation which permits natural language interface and runs interactively with videotape to be fair to good (K = .61).

Computer-Assisted Instruction↗
PubMed1991

Evaluation of SAPHIRE: an automated approach to indexing and retrieving medical literature.

An analysis of SAPHIRE, an experimental information retrieval system featuring automated indexing and natural language retrieval, was performed on MEDLINE references using data previously generated for a MEDLINE evaluation. Compared with searches performed by novice and expert physicians using MEDLINE, SAPHIRE achieved comparable recall and precision. While its combined recall and precision performance did not equal the level of librarians, SAPHIRE did achieve a significantly higher level of absolute recall. SAPHIRE has other potential advantages over existing MEDLINE systems. Its natural language interface does not require knowledge of MeSH, and it provides relevance ranking of retrieved references.

Abstracting and Indexing↗
PubMed1991

Object-oriented controlled-vocabulary translator using TRANSOFT + HyperPAD.

Automated coding of surgical pathology reports is demonstrated. This public-domain translation software operates on surgical pathology files, extracting diagnoses and assigning codes in a controlled medical vocabulary, such as SNOMED. Context-sensitive translation algorithms are employed, and syntactically correct diagnostic items are produced that are matched with controlled vocabulary. English-language surgical pathology reports, accessioned over one year at the Baltimore Veterans Affairs Medical Center, were translated. With an interface to a larger hospital information system, all natural language pathology reports are automatically rendered as topography and morphology codes. This translator frees the pathologist from the time-intensive task of personally coding each report, and may be used to flag certain diagnostic categories that require specific quality assurance actions.

Algorithms↗
PubMed1990

[Registration of detailed data in the medical record or how to translate "impressions" into measurable observations].

Every medical case record represents a mass of data (texts, pictures, figures, etc.) in an unstructured form. The physician needs to retrieve this data via several access routes: temporal (dependent on date or sequences of events), type of data (diagnostic, treatment, clinical signs, laboratory findings, image descriptors, all with their interrelationships), or depending on the severity of the disease, etc. Retrieval of this data fulfils several functions: circulation of a case record among specialists, assistance in summarizing a long and complex clinical course, comparison of patients, research, and teaching. Three projects are described which have the same aim: structuring of the case record in order to retrieve detailed data on patients as individuals and describe clinical courses on the basis of measurable observations. This structure must be understandable to a computer (directly or indirectly) so that searches and comparisons can be performed automatically. The first project, entitled "indexed paragraph prototype" reproduces the structure of the problem-oriented case record and is designed to input the Medical Outpatients Department's follow-up notes into the computer. The second, "automatic language analysis", aims to exploit two characteristics of medical language, its omnipresence in the case record and its reliability, in view of its status as the spontaneous vehicle of communication between physicians. The third, "collection of clinical signs during consultation", is based on a prospective collection of all elements of clinical observation, structured temporally consultation by consultation. The purpose of precise collection of detailed and measurable observations in individual patients is to identify those among the clinical signs which display the greatest power of discrimination, i.e. those which best serve to predict the case's evolution.

Abstracting and Indexing↗
PubMed1990

[Comparison of 2 series of autopsies observed at Johns-Hopkins Medical Center, Baltimore (JHMI) and at the Neuchâtel Institute of Pathology (INAP)].

We are reporting the first results of a comparative study of 100 consecutive autopsies and their clinical diagnoses, observed at the Johns Hopkins Medical Institutions (JHMI) and at the Institut neuchâtelois d'anatomie pathologique (INAP). The diagnoses of the two series were coded according to the two different systems used currently at the two institutions. The data from Baltimore were automatically classified by a special "key word method" using the categories of the Index Medicus (MeSH = Medical subject headings). We proceeded then to a second recording by SNOMED codes, introduced into the computer system in the same way as we document the autopsy diagnoses in Neuchâtel. The two series could be compared in detail according to topographical, morphological and aetiological parameters. The over-all repartition of the examined cases shows a higher incidence of newborns in Baltimore (23), in Neuchâtel we observed only 7 newborn autopsies. The mean age was inferior in Baltimore (males: 53.5 years for JHMI, 73.1 years for INAP; females: 58.4 years for JHMI, 66.2 years for INAP). The number of diagnoses per autopsy was 31.9 at JHMI, and 50.1 in Neuchâtel. The topographical distribution of clinical and autopsy diagnoses showed a higher frequency of central nervous system lesions in JHMI which might be explained by the activity of a neuropathological division. Findings concerning the morphological categories revealed a higher frequency in JHMI for traumatic abnormalities (7.2% vs 2.5%), malformations (4.3% vs 0.8%), whereas inflammation and fibrosis and degenerative lesions were more often encountered in Neuchâtel. The differences in morphological observations could be attributed to a higher proportion of newborn cases in JHMI with complex malformation syndromes.(ABSTRACT TRUNCATED AT 250 WORDS)

Adolescent↗
PubMed1989

[RGSS-IDJ and its application to cranial computed tomography].

RGSS-IDJ is developed as the Japanese version of Report Generation Support System for Imaging Diagnosis (RGSS-ID), which is a developmental computer system that applies artificial intelligence (AI) methods to a reporting system. Now RGSS-IDJ supports the report generation of cranial computed tomography. A representation scheme called Generalized Finding Representation (GFR) is proposed, to bridge the gap between natural language expressions in the radiographic report and AI methods. GRF for RGSS-IDJ is the same as for RGSS-ID. The basic style for entering the findings on the radiograph is the dialogue system with the routine of query and answering it by selecting items with a mouse. This system encodes the input findings into the network expressions, which are represented as the list form in the LISP computer language. And, it reserves them into the knowledge data base. The content of the report will be able to be utilized for various analyses within AI paradigm. The final radiographic report is made in the natural Japanese language.

Artificial Intelligence↗
PubMed1989

Generation of surgical pathology report using a 5,000-word speech recognizer.

Pressures to decrease both turnaround time and operating costs simultaneously have placed conflicting demands on traditional forms of medical transcription. The new technology of voice recognition extends the promise of enabling the pathologist or other medical professional to dictate a correct report and have it printed and/or transmitted to a database immediately. The usefulness of voice recognition systems depends on several factors, including ease of use, reliability, speed, and accuracy. These in turn depend on the general underlying design of the systems and inclusion in the systems of a specific knowledge base appropriate for each application. Development of a good knowledge base requires close collaboration between a domain expert and a knowledge engineer with expertise in voice recognition. The authors have recently completed a knowledge base for surgical pathology using the Kurzweil VoiceReport 5,000-word system.

Artificial Intelligence↗
PubMed1988

Integrated pathology reporting, indexing, and retrieval system using natural language diagnoses.

Pathology computer systems are making increasing use of natural language diagnoses. The Johns Hopkins Medical Institutions integrated pathology reporting system, a commercial product with extensive, locally added enhancements, covers all information management functions within autopsy and surgical pathology divisions and has on-line linkages to clinical laboratory reports and the medical library's Mini-MEDLINE system. All diagnoses are written in natural language, using a word processor and spelling checker. A security system with personal passwords and different levels of access for different staff members allows reports to be signed out with an electronic signature. The system produces financial reports, overdue case reports, and Boolean searches of the database. Our experience with 128,790 consecutively entered pathology reports suggests that the greater precision of natural language diagnoses makes them the most suitable vehicle for follow-up, retrieval, and systems development functions in pathology.

Artificial Intelligence↗
PubMedDate not supplied

[Methodology for the development of expert systems of viral epidemiology].

The proposed methodology for the elaboration of the base of knowledge uses a tree of the hierarchical entities and a simplified variant of the natural language. The resolution system is based on an extension of the predicate calculation containing, in an explicit way, entities of different nature, and among these the correlations giving the rules of deduction.

Artificial Intelligence↗
PubMed1986

Studies in artificial aphasia: experiments in processing change.

Computational neurolinguistics, an integrated approach to cognitive modelling of neural processes which may subserve natural language performance, attempts to build computational models that model behavior on two levels: at the neural process level and at the human performance level in its normal state and under pathological conditions. HOPE is one example of such a model. It demonstrates how the design and implementation of such models can provide insights into how a brain-like architecture can produce a behavior as complex as natural language. This paper will briefly describe the neurally motivated or 'natural computation' processes which produce the model's observable and verifiable behavioral results. Experiments with artificially induced aphasia on HOPE will then be described, showing that the results of simulation produce hypothesized patient profiles that are unique. These profiles illustrate the suggested contribution of the computational neurolinguistics research approach as a tool for investigating the breakdown of language performance and its potential contribution to understanding brain function.

Aphasia↗
PubMed1987

Selective word-learning deficits in aphasia.

The purpose of this study was to determine whether patients with focal brain damage can learn anything about a new word, and if so, whether selected aspects of the new word are acquired depending on the nature of the patient's language processing deficit. In the context of drawing pictures with a number of felt pens identical in all respects except color, agrammatic Broca's aphasics and fluent aphasics were exposed to the new word "bice," an adjective referring to the dark green portion of the color spectrum. Our findings revealed that brain-damaged patients can engage in lexical acquisition, but Broca's aphasics and fluent aphasics apparently learn about different aspects of the new word. Specifically, on successive exposures to "bice," both Broca's aphasics and fluent aphasics exhibit progressively more accurate hypotheses for identifying a bice-colored object. During subsequent assessments, agrammatic aphasics reveal on a metalinguistic judgment task their significant difficulty appreciating the grammatical form class of "bice"; on an object classification task, fluent aphasics are significantly impaired in their classification of bice-colored objects as "bice." Taken together, these unique word-learning profiles reinforce earlier observation on the selective nature of language processing deficits after focal left-hemisphere insult, and support the claim that separate language processing devices may be selectively compromised in different groups of aphasics. The relationship between syntactic and lexical semantic processing is discussed.

Adult↗
PubMed1988

Computer-assisted assessment of patient care in the hospital.

Using his natural language medical text analyzing system, the author has computer-processed the discharge summary segment of the patient record. The output of the text analyzer is a list of medical facts. The low-cost, high productivity process is eminently suited for screening the quality of clinical care provided in the hospital.

Artificial Intelligence↗
PubMedDate not supplied

On the role of perception in shaping phonological assimilation rules.

Assimilation of nasals to the place of articulation of following consonants is a common and natural process among the world's languages. Recent phonological theory attributes this naturalness to the postulated geometry of articulatory features and the notion of spreading (McCarthy, 1988). Others view assimilation as a result of perception (Ohala, 1990), or as perceptually tolerated articulatory simplification (Kohler, 1990). Kohler notes that certain consonant classes (such as nasals and stops) are more likely than other classes (such as fricatives) to undergo place assimilation to a following consonant. To explain this pattern, he proposes that assimilation tends not to occur when the members of a consonant class are relatively distinctive perceptually, such that their articulatory reduction would be particularly salient. This explanation, of course, presupposes that the stops and nasals which undergo place assimilation are less distinctive than fricatives, which tend not to assimilate. We report experimental results that confirm Kohler's perceptual assumption: In the context of a following word initial stop, fricatives were less confusable than nasals or unreleased stops. We conclude, in agreement with Ohala and Kohler, that perceptual factors are likely to shape phonological assimilation rules.

Female↗