Search PubMed⌕ Search

Biomedical subjects

P A Michel

Publications and source records attributed to P A Michel.

14 recordsLinked to original sources
PubMed1998

Full text multilingual automatic morphosemantems for stand-alone or Internet based applications.

The authors present an automatic tool able to provide real-time morphosemantic decomposition of natural language sentences in French, German and English. This tool demonstrates the feasibility of Natural Language Processing on standard PC computers and the technology involved has been successfully implemented in daily used applications in several European hospitals. It considerably alleviates the burden of coding with various international classification and enhances the quality of the final results. This tool, delivered on PC platforms, is highly convivial and provides a versatile interface to any existing applications based on the Microsoft Windows standards. Moreover, all high levels functions have been encapsulated in Object Oriented Components and can therefore be reused using the Common Object Model standards to develop stand-alone or Internet applications.

Classification↗
PubMed1998

Automatic extraction of linguistic knowledge from an international classification.

Automatic extraction of knowledge from large corpus of texts is an essential step toward linguistic knowledge acquisition in the medical domain. The current situation shows a lack of computer-readable large medical lexicons, with a partial exception for the English language. Moreover, multilingual lexicons with versatility for multiple languages applications are far from reach as long as only manual extraction is considered. Computer-assisted linguistic knowledge acquisition is a must. A multilingual lexicon differs from a monolingual one by the necessity to bridge the words in different languages. A kind of interlingua has to be built under the form of concepts to which the specific entries are attached. In the present approach, the authors have developed an intelligent rule-based tool in order to focus on a multilingual source of medical knowledge, like the International Classification of Disease (ICD) which contains a vocabulary of some 20,000 words, translated in numerous languages.

Disease↗
PubMed1996

Knowledge sources for Natural Language Processing.

This paper aims at reviewing the problem of feeding Natural Language Processing (NLP) tools with convenient linguistic knowledge in the medical domain. A syntactic approach lacks the potential to solve a number of typical situations with ambiguities and is clearly insufficient for quality treatment of natural language. On the other hand, a conceptual approach relies on some modelling of the domain, of which the elaboration is d long-term process and where the ultimate solutions are far from being recognised and universally accepted. In-between is the beauty of the compromise. How can we significantly improve the coverage of linguistic knowledge in the years to come?

Artificial Intelligence↗
PubMed1995

Multilingual natural language generation as part of a medical terminology server.

Re-usable and sharable, and therefore language-independent concept models are of increasing importance in the medical domain. The GALEN project (Generalized Architecture for Languages Encyclopedias and Nomenclatures in Medicine) aims at developing language-independent concept representation systems as the foundations for the next generation of multilingual coding systems. For use within clinical applications, the content of the model has to be mapped to natural language. A so-called Multilingual Information Module (MM) establishes the link between the language-independent concept model and different natural languages. This text generation software must be versatile enough to cope at the same time with different languages and with different parts of a compositional model. It has to meet, on the one hand, the properties of the language as used in the medical domain and, on the other hand, the specific characteristics of the underlying model and its representation formalism. We propose a semantic-oriented approach to natural language generation that is based on linguistic annotations to a concept model. This approach is realized as an integral part of a Terminology Server, built around the concept model and offering different terminological services for clinical applications.

Language↗
PubMed1995

Toward a medical linguistic knowledge base.

This paper presents the design of a Medical Linguistic Knowledge Base (MLKB). This MLKB is intended to be the multilingual recipient for all the declarative knowledge about languages. It includes words, their syntax and their conceptual representation, typology of concepts of the domain, rules for semantic analysis and conceptual schemata. For that purpose, Sowa's conceptual graphs are considered as an adequate knowledge representation. The MLKB will be an enormous body of information, and the difficulty to feed it and to validate it appears immediately. Therefore, it is necessary to start an international initiative to merge efforts from different groups.

Language↗
PubMed1995

Word segmentation processing: a way to exponentially extend medical dictionaries.

One of the most critical problems of automatic natural language processing (NLP) is the size of the medical lexicons. The set of compound medical words and the continual creation of new terms renders medical lexicons exhaustive beyond question. The structure of such dictionaries usually consists of two parts: 1) the morphological and sometimes syntactical information necessary to identify, on a grapheme level, a given word in a sentence, and 2) the part often devoted to conceptual knowledge associated with the recognized word. It is only when these two prerequisites are fulfilled that an attempt to understand the meaning of a whole expression is possible. The approach developed in this paper is a pragmatic way to rapidly increase the lexico-semantic part of medical dictionaries. We developed a semi-automatic tool, as a prototype to demonstrate the feasibility of this approach. This tool is able to translate almost any diagnosis expressed in French into its equivalent in the ICD-9CM coding scheme.

Dictionaries, Medical as Topic↗
PubMed1995

LUCID: a semi-automated ICD-9 encoding system.

The natural language approach to diagnosis encoding will certainly become a widespread technique during the second half of this decade. Accessing standard codes by numbers and keywords will be more and more considered a loss of time and efficiency. We present a demonstration of a natural language based encoding system for ICD, called LUCID, which considerably alleviates the burden of coding with ICD classification and enhances the quality of the encoded list of diagnoses. This tool, delivered on a PC platform, is very convivial and provides a versatile interface to any existing application based on Microsoft Windows standards.

Abstracting and Indexing↗
PubMed1995

Current trends with natural language processing.

Natural Language Processing in the medical domain becomes more and more powerful, efficient, and ready to be used in daily practice. The needs for such tools are enormous in the medical field, due to the vast amount of written texts for medical records. In the authors' point of view, the Electronic Patient Record (EPR) is achieved neither with Information Systems of all kinds nor with commercially available word processing systems. Natural Language Processing (NLP) is one dimension of the EPR, as well as Image Processing and Decision Support Systems. Analysis of medical texts to facilitate indexing and retrieval is well known. The need for a generation tool is to produce progress notes from menu driven systems. The computer systems of tomorrow cannot miss any single dimension. Since 1988, we've been developing an NLP system; it is supported by the European program AIM (Advanced Informatics in Medicine) within the GALEN and HELIOS consortium and the CERS (Commission d'Encouragement á la Recherche Scientifique) in Switzerland. The main directions of development are: a medical language analyzer, a language generator, a query processor, and dictionary building tools to support the Medical Linguistic Knowledge Base (MLKB). The knowledge representation schema is essentially based on Sowa's conceptual graphs, and the MLKB is multilingual from its design phase; it currently incorporates the English and the French languages; it will also continue using German. The goal of this demonstration is to provide evidence of what exists today, what will be soon available, and what is planned for the long term. Complete sentences will be processed in real time, and the browsing capabilities of the MLKB will be exercised. In particular, the following features will be presented: Analysis of complete sentences with verbs and relatives, as extracted from clinical narratives, with special attention to the method of "proximity processing" as developed in our group and the rule based approach to language description to resolve the specific surface language problems as well as the language independent semantic situations. Comparison of results for English, French, and German sentences, showing the commonalities between these languages and, therefore, the re-usable features and the language specific aspects. Generation of noun phrases in English and French, showing the opportunities for translation between these two languages. Application of the analyzer to build a knowledge representation of ICD under the form of conceptual graphs and presentation of the possibilities of a natural language encoding of diagnosis. Strategies for query processing through a sample of abdominal ultrasonography reports, which have been analyzed and stored under the form of conceptual graphs. Feeding in and browsing of the Medical Linguistic Knowledge Base and other Dictionary Building Tools, using the perspective of an international initiative to converge towards a multilingual universal solution, valid for the medical domain. The demonstration platform is Microsoft Windows 4 on a PC, with Microsoft Visual Basic as the GUI and Quintus Prolog as NLP tools language. The same programs were originally developed for Unix-based workstations and are available on multiple platforms under Motif and X11. .

European Union↗
PubMed1995

Representing clinical narratives using conceptual graphs.

The analysis of medical narratives and the generation of natural language expressions are strongly dependent on the existence of an adequate representation language. Such a language has to be expressive enough in order to handle the complexity of human reasoning in the domain. Sowa's Conceptual Graphs (CG) are an answer, and this paper presents a multilingual implementation, using French, English and German. Current developments demonstrate the feasibility of an approach to natural Language Understanding where semantic aspects are dominant, in contrast to syntax driven methods. The basic idea is to aggregate blocks of words according to semantic compatibility rules, following a method called Proximity Processing. The CG representation is gradually built, starting from single words in a semantic lexicon, to finally give a complete representation of the sentence under the form of a single CG. The process is dependent on specific rules of the medical domain, and for this reason is largely controlled by the declarative knowledge of the medical Linguistic Knowledge Base.

Artificial Intelligence↗
PubMed1994

ARTEMIS-2: an application development experiment with the HELIOS environment.

A medical application is a highly complex system that embraces many data types and a very large number of data processing functions and methods. The development of integrated software engineering environments has deeply changed the conception of applications and the profile of the application developers. In this paper, we address the problem of the development process of a specific multimedia application, called ARTEMIS, within the distributed HELIOS environment. The application is intended to manage information about hypertensive patients, in particular, retrieval and display of administrative, clinical and biological data and display and analysis of digital angiography images and medical reports. The objective is to show how the developer can use, customize and organize the services HELIOS provides. A particular focus is set on reuse strategies and integration during the development process. A scenario has been realized and illustrates the current state of the application. The discussion focuses on the advantages of such distributed environments in medical application development.

Computer Communication Networks↗
PubMed1994

Natural language processing of medical texts within the HELIOS environment.

A large number of hospital applications are potentially interested in natural language processing since they currently heavily depend on an efficient use of a huge amount of textual information. The need for systems that are able to accept multiple European languages is of paramount interest, as language barriers can be a strong impediment for large-scale communication in Europe, in particular regarding telemedicine. In the context of the AIM project HELIOS, the Natural Language Processing (NLP) component offers a large variety of medical services according to natural language free input. It allows the multilingual analysis of medical texts (currently in English, French and German) and the storage of the meaning of these texts under a deep knowledge representation that can be queried whenever it is needed. In addition, it provides facilities to handle knowledge source embedded into the conceptual typologies and into the dictionaries. This article aims at describing all these functionalities and their integration into the environment of the HELIOS project.

Artificial Intelligence↗
PubMedDate not supplied

Medical dictionaries for patient encoding systems: a methodology.

Medical language is highly compositional and makes extensive use of common roots, especially Latino-Greek roots. Besides words devoted to common sense, medical language presents some typical characteristics, especially on morphological and semantic aspects of word formation. Morphological decomposition and identification precedes semantic analysis. It is only when these two prerequisites are fulfilled that an attempt to grasp the meaning of a whole expression is made possible. The main aim of the proposed approach is that of coping with 'the lack of coverage of the medical lexical knowledge', in order to help physicians find the correct international classification for diseases (ICD) codes for a written diagnosis. The proposed methodology allows the development of a powerful dynamic dictionary dedicated to natural language processing in the field of diagnoses and narrative procedures. It describes the design of an analyser that can profit from a dictionary. The methods used have proved to be efficient for various classifications, s well as for multiple languages, as the system presently supports French, German, English and Dutch for ICD-9 and ICD-10 classifications.

Classification↗