Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “artificial intelligence”

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 1,279 records · Page 71Linked to original sources
PubMed1986

The computer as a tool for learning.

Experimenters from the beginning recognized the advantages computers might offer in medical education. Several medical schools have gained experience in such programs in automated instruction. Television images and graphic display combined with computer control and user interaction are effective for teaching problem solving. The National Board of Medical Examiners has developed patient-case simulation for examining clinical skills, and the National Library of Medicine has experimented with combining media. Advances from the field of artificial intelligence and the availability of increasingly powerful microcomputers at lower cost will aid further development. Computers will likely affect existing educational methods, adding new capabilities to laboratory exercises, to self-assessment and to continuing education.

Computer-Assisted Instruction↗
PubMed1987

[An expert system of aiding decision making in breast pathology connected to a clinical data base].

The René Huguenin Cancer Center holds a medical file for each patient which is intended to store and process medical data. Since 1970, we introduced computerization: a development plan was elaborated and simultaneously a statistical software (Clotilde--GSI/CFRO) was selected. Thus, we now have access to a large database, structured according to medical rationale, and utilizable with methods of artificial intelligence towards three objectives: improved data acquisition, decision making and exploitation. The first application was to breast pathology, which represents one of the Center's primary activities. The structure of the data concerning patients is by all criteria part of the medical knowledge. This information needs to be presented as well as processed with a suitable language. To this end, we chose a language-oriented object, Mering II, usable with Apple and IBM 4 micro-computers. This project has already allowed to work out an operational model.

Breast Neoplasms↗
PubMed1985

Laboratory-clinician interaction and the interpretation of test results.

Medical care has become more complex and dependent on technological innovations such as computerised tomography and laser surgery. The use of computers for diagnostic purposes and the emergency of the field of artificial intelligence as purveyor of the medical data base, have given the practitioners a very large body of data to which they must apply their considerable ability. However, the more methods become sophisticated and technically advanced, the greater the need for close cooperation and sharing of expertise between the various components of the delivery of health services. In this context, the clinical laboratory is an amplification of the physician's senses so that he can 'see', 'hear' and 'feel' at cellular, molecular and atomic levels. This is carried out by optimised use of laboratory capabilities by the clinical biochemist and the physician. The bedside is the meeting point between the clinician and his consultants and the patient. It is here that a decision will be made that will have a profound effect on the health of the patient. The ability of the clinical biochemist to design test strategies, control sample handling and interpret laboratory data, is greatly enhanced by having him at the bedside, in close cooperation with the physician. This relationship between the patient, the physician and the laboratory is seen in Figure 6.2. The physician, even with the most sophisticated and modern bedside instruments, cannot be expected to be an expert in properly evaluating test results. He is an expert in synthesising a large body of expert evidence into diagnosis and treatment of his patient. Refusing to consult the clinical biochemist on the interpretation of laboratory data or receiving non-expert consultation can have serious consequences to the patient. However, the clinical biochemist can only truly provide expert analysis if he is 'near' the patient and can be attuned to his physical as well as his emotional and social state which have a great influence on the interpretation of his physiological status. This latter aspect was extremely well put by Peabody almost 60 years ago and is still correct today. Clinical biochemistry must be 'near' the patient to fully realise its professional status.

Blood Specimen Collection↗
PubMed1985

Expert systems in medicine: a biomedical engineering perspective.

Knowledge-based expert systems for medical applications have received considerable attention in recent years. In this review, fundamental terms and notions of artificial intelligence techniques as applied to expert systems are introduced. The most well-known and influential medical expert systems are discussed in detail, and newer efforts are surveyed. A critical comparison of strengths and weaknesses of the systems is made, discussing depth and complexity of knowledge, acquisition of knowledge, user interaction and explanations, knowledge engineering tools, system evaluations, and user resistance. Long- and short-term trends are appraised.

Cholestasis↗
PubMed1985

[Presentation of a system for computer-aided decision in diabetes therapeutics].

This paper presents Sphinx, an expert system for computer-aided diagnosis in diabetes therapeutic. This interactive program makes logic deductions, comprehensible for users and it is not necessary for them to know artificial intelligence methodology. An improvement of the first results is possible and necessary, but the reliability of the advice already expressed by the system makes it useful as a future physician's implement.

Computers↗
PubMed1983

Computer realization of some decision process in neurological diagnostics.

A digital computer-assisted system is described which enables acquisition of data on an objective neurologic finding and results of the cerebellum examinations. On the basis of the registered finding the system provides full automated topographic or syndromological diagnosis. It avails of artificial intelligence and operates the data base of concrete medical knowledge using logical decision making rules equivalent with diagnostic views of a neurologist.

Computers↗
PubMed1984

Quantitative theories for clinical decision making.

Clinical work involves a great many decisions in situations characterized by uncertainty and by the need for value judgments. Formal, mostly quantitative approaches to medical decision making have been developed over the last thirty years. The best known theories are discriminant analysis or supervised pattern recognition, decision analysis and, more recently, artificial intelligence. Discriminant analysis is used for estimating diagnostic probabilities on the basis of diagnostic data. Outcome, action, state, probability, utility, and choice are the main concepts of decision analysis. Problems of diagnosis and therapy choice can be structured within the framework of decision analysis. The quality of a decision analysis largely depends on a good structuring of the problem, and on the quality of the probability and utility assessments made during the analysis. Decision analysis is basically a prescriptive theory for decision making in individual patient management. Other areas of application include basic and advanced teaching, formulation, explication and analysis of clinical problems, detection of gaps in clinical knowledge, development of clinical protocols, etc.

Decision Making↗
PubMed1983

Homologous partial sequences in dehydrogenases.

Such details of the primary structure were sought that are common in all dehydrogenases of known amino acid sequence. Twenty-six sequences of eight kinds of dehydrogenase (D-glyceraldehyde-3-phosphate dehydrogenase, alcohol dehydrogenase, lactate dehydrogenase, glutamate dehydrogenase, glycerol-3-phosphate dehydrogenase, ribitol dehydrogenase, L-hydroxyacyl-CoA dehydrogenase and homoserine dehydrogenase) have been compared by the aid of the artificial intelligence language Prolog, the amino acids being classified into groups according to their chemical properties, and alpha-helix or beta-sheet-forming abilities. We found tetrapeptides that occurred in all dehydrogenases examined. By using these tetrapeptides as markers a population of 84 partial sequences has been described. The partial sequences constituting this population are peptides comprising 35 residues. It has been shown statistically that these peptides form a homogeneous sample as regards the frequency of occurrence of amino acid groups. This statistically homogeneous partial sequences can be regarded as homologous and it is assumed that their presence is characteristic of dehydrogenases.

Amino Acid Sequence↗
PubMedDate not supplied

Model-driven visualization of coronary arteries.

In a joint project between the Department of Computer and Information Sciences and the Department of Radiology, we are applying techniques of artificial intelligence to improve clinical performance in coronary arteries. Specifically, we are investigating how images from intravenous digital subtraction angiography (DSA) can be enhanced so that their efficacy for lesion detection and quantitation becomes comparable with that of the more dangerous procedure of selective coronary arteriography. The enhancement techniques (which include algorithms for 3-dimensional vessel detection, reconstruction and display, as well as for accurate lumen-size estimation) are based on models of (i) the 3-dimensional topological structure of the coronary arterial tree, (ii) myocardial dynamics, and (iii) the X-ray imaging process involved in producing digital subtraction angiograms. The evaluation of these model-driven visualization techniques is done by the standard psychophysical method of Receiver Operating Characteristic (ROC) analysis applied to observer performance tests on images from an animal coronary atherosclerosis model.

Analog-Digital Conversion↗
PubMed1982

DPS--a computerised diagnostic problem solver.

The paper contains a short description of the DPS system which is a computerized diagnostic problem solver. The system is under development of the Research Institute of Medical Bionics in Bratislava, Czechoslovakia. Its underlying philosophy yields from viewing the diagnostic process as process of cognitive problem solving. The implementation of the system is based on the methods of Artificial Intelligence and utilisation of production systems and frame theory should be noted in this context. Finally a list of program modules and their characterisation is presented.

Diagnosis, Computer-Assisted↗
PubMed1982

The microcomputer in medical instrumentation. Aspects of an automated monitoring system.

The major technical trends in medical computing include sensor devices and artificial intelligence techniques. Use of the computer will render instruments more accessible and make them friendlier to the user. Examples include automation of gain setting in echocardiology and instant computation of M-mode data. In intensive care units, computers have become an integral part of patient monitoring over the last two decades. Third generation microcomputer-based systems are presently helping to solve many earlier problems. Features such as touch-sensitive keys or screens (eliminating keyboards) and integrated front-end are greatly increasing user acceptance. A new monitoring system developed at the Thorax Center for intensive cardiac patient monitoring and provided with these features will be described.

Computers↗
PubMed1995

[Contemporary strategies and methods of modeling antiparasitic drugs].

Contemporary methods of directed chemotherapy are based on multi-step procedures, which require co-ordinated activities of interdisciplinary teams of biochemists, pharmacologists, geneticists, crystallographers as well as computer scientists. Biochemists select the proper target, such as an enzyme, throughout screening of the biochemical influence of compounds-potential drugs on this target. For further research they use targets with very low inhibition constants (> 10(-6) M). Determination of the relation between therapeutic activity of the compound and modelling of its chemical structure constitutes an important part of the procedure. The most important part of the procedure is the recognition of the primary structure of the target. The two following pathways allow to do that: 1. isolation of DNA and gDNA or cDNA-started cloning of a gene responsible for production of the target protein and then its sequencing. 2. purification and crystallization of the target protein and further computer-aided processing of crystallographic data in order to determine the primary structure. Computational chemistry (C/C) methods are the basic part of the procedure of molecular modelling (M/M) of a target molecule and its interactions with a molecule of the future drug. Data obtained using a technology which engages the C/C and M/M methods not only allow to determine the aminoacid sequence of the target protein in question (e.g. a unique parasite enzyme); they also enable to further speculate on its secondary and tertiary structures. Such structure includes specified number of repeated motifs of alpha-helixes, beta-sheets and loops or turns. Particularly, the "barrel" structure is very common in numerous enzymes. Two following examples of research on target-antiparasitic drug interactions is presented. They are the interaction between phosphoglicerate kinase in Leishmania and drug suramin and malic enzyme of Trichinella and drug closantel. New promising targets for new anti-protozoan drugs (protozoa of Trypanosoma species) include e.g. microbody translocation signal in kinetosom proteins (SKL) or protein blocking the transport of proteins to glycosomes-metabolic centres in Trypanosoma (repetitive groups of QRLQ). Recently, scientists from Arris Pharmaceutical (San Francisco) have considered, employing new data, up to 100 to fully characterize the surface structure of a molecule, using the systems of artificial intelligence.

Anthelmintics↗
PubMed1995

HOROPLAN: computer-assisted nurse scheduling using constraint-based programming.

Nurse scheduling is a difficult and time consuming task. The schedule has to determine the day to day shift assignments of each nurse for a specified period of time in a way that satisfies the given requirements as much as possible, taking into account the wishes of nurses as closely as possible. This paper presents a constraint-based, artificial intelligence approach by describing a prototype implementation developed with the Charme language and the first results of its use in the Rouen University Hospital. Horoplan implements a non-cyclical constraint-based scheduling, using some heuristics. Four levels of constraints were defined to give a maximum of flexibility: French level (e.g. number of worked hours in a year), hospital level (e.g. specific day-off), department level (e.g. specific shift) and care unit level (e.g. specific pattern for week-ends). Some constraints must always be verified and can not be overruled and some constraints can be overruled at a certain cost. Rescheduling is possible at any time specially in case of an unscheduled absence.

France↗
PubMed1993

[Informatics in food microbiology: applications for the identification of microorganisms].

Recent progress in analytical procedures (chemotaxonomy, molecular biology) and its contribution to the development of microbial Systematics and Identification have been examined. The benefit that these techniques have gained from the technological progression experienced by computer hardware and software (artificial intelligence and numerical taxonomy) and the improvement in tools for the study of characteristics and interrelationships between microorganism are also reported.

Bacteria↗
PubMed1993

[General aspects of the diagnostic process in acute organic psychoses].

Following the Central European tradition of the triad system of psychiatry of Kurt Schneider (I. Abnormal Reactions/Personalities, II. Acute/Chronic Organic Psychosis, III. Schizophrenia/Cyclothymia) (19) a series of 210 cases of acute organic psychoses was collected. In contrast to DSM-III and in accord with W.A. Lishman (10), this series comprises not only delirium, but the entire range of impairments of consciousness from the "Durchgangssyndrom" of H.H. Wieck (24) to clouding of consciousness and coma, twilight states or apallic syndromes. From this series we present some elementary quantitative data concerning the age distribution of different etiologies, that are generally considered helpful in the diagnostic process. In contrast to this opinion, a single case of bromazepam-induced delirium shows: The anamnesis may be misleading, the general and neurological examination as well as the radiological, laboratory and electrophysiological tests could be insignificant. The essential diagnostic tool is the descriptive or phenomenological psychopathological observation. Compared to it, the general data of the age distribution of different etiologies of organic psychoses are of comparably little help in making a diagnosis. Artificial intelligence and medical expert systems are set out to replace the diagnosis of the physician. Eliminative materialistic neuroscience is set out to replace old-fashioned descriptive psychopathology by quantitative electrical and chemical data. Through the method of systematically confronting general quantitative data with suitably chosen single cases it should be possible to find out essential differences between general reductionist statements and the properties of complex qualitative phenomena like the individual human mind.

Adolescent↗
PubMed1994

The management of information: storage and retrieval of data.

Internationally harmonized and cost-effective control of chemicals marketed worldwide greatly depend both on the generation of and easy access to reliable and comparable experimental information. Stored data are of use only if information can be retrieved quickly in an understandable form. Some models and theories of information retrieval (e.g. fuzzy set theory, probabilistic approach, artificial intelligence) are briefly discussed first, then followed by applications (such as indexing and clustering techniques). Finally the structure of databases is briefly reviewed.

Database Management Systems↗
PubMed1995

Neural network classification of corneal topography. Preliminary demonstration.

PURPOSE: Videokeratography is a powerful tool for the diagnosis of corneal shape abnormalities. However, interpretation of the topographic map is sometimes difficult, especially when pathologies with similar topographic patterns are suspected. The neural networks model, an artificial intelligence approach, was applied for automated pattern interpretation in corneal topography, and its usefulness was assessed. METHODS: One hundred eighty-three topographic maps were selected and classified by human experts into seven categories: normal, with-the-rule astigmatism, keratoconus (mild, moderate, advanced), postphotorefractive keratectomy, and postkeratoplasty. The maps were divided into a training set (108 maps) and a test set (75 maps). For each map, 11 topography-characterizing indices calculated from the data provided by the TMS-1 videokeratoscope, plus the corresponding diagnosis category, were used to train a neural network. RESULTS: The correct classification was achieved by a trained neural network for all 108 maps in the training set. In the test set, the neural network correctly classified 60 of 75 maps (80%). For every category, accuracy and specificity were greater than 90%, whereas sensitivity ranged from 44% to 100%. CONCLUSIONS: With further testing and refinement, the neural networks paradigm for computer-assisted interpretation or objective classification of videokeratography may become a useful tool to aid the clinician in the diagnosis of corneal topographic abnormalities.

Cornea↗