SEARCH · PubMed Health

Results for “Expert Systems”

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 847 records · Page 47Linked to original sources

PubMed1995

[Integrated gait analysis: a complex expert system approach to detailed data evaluation].

Integrated Gait Analysis is still not accepted as an important tool in a routine clinical environment because of its lack of assessment procedures closely based on the raw data measurements. Furthermore, an adequate treatment of the complex process of human movement is missing. This contribution is intended to successively describe the scheme of a dedicated approach for gait analysis data assessment. Therein, different technologies already discussed in the literature have been used. Some examples are given for the purpose of illustration. To provide in advance sufficiently different and well-known measurement conditions, extreme manipulations in the alignment of below-knee prostheses have been carried out. It is shown that some specific data formats consisting of subgroups of gait patterns have turned out to be important for further partial processing and for interfacing the different subsystems. However, due to data characteristics of the raw data and the lack of a well-founded a priori knowledge, some principal questions have to be left open. Therefore, on the basis of the different results presented below and their discussion, a more detailed model of the process is proposed for further research following this concept.

Artificial Limbs↗

PubMed1998

Expert systems for the interpretation of serum proteins.

The ready availability of inexpensive desk-top computers with enormous disk storage has made the practicality of computer assisted medical interpretive software a reality. There seems little question that these programs could be of enormous help to physicians. However, there are daunting problems to their creation, including the lack of standards for clinical diagnostic precision or accuracy and paucity of helpful literature. As a result, the final products may be quite different. Little effort has been devoted in the laboratory to produce programs which could have great benefit in bridging the gap between laboratorians and clinicians. In a few circumscribed areas where the interpretation of laboratory measurements have been well studied in relation to patient demographics and to the final outcome, the impact has been enormous. The prime example is prenatal diagnosis of neural tube defects, and certain chromosomal and developmental abnormalities. Viewed as an obstacle by most people able and willing to attempt to create such programs is the omnipresence of necessary regulation. A brief overview of the general structure of a program to assist with the interpretation of serum proteins is given as a model in the perspective of current knowledge and state of the relevant literature.

Blood Proteins↗

PubMed1990

Identification of rodent carcinogens by an expert system.

CASE, an artificial intelligence method for identifying structural determinants responsible for biological activity was applied to the U.S. National Toxicology Program (NTP) cancer bioassay results. CASE identified structures which were significantly associated with rodent carcinogenicity. On the basis of these structural determinants CASE exhibited a sensitivity of 0.98 and a specificity of 1.00. CASE showed a similarly remarkable performance in predicting the carcinogenicity, or lack thereof, of chemicals not in the NTP data base. A comparison between the activating structures (biophores) responsible for mutagenicity in Salmonella and rodent carcinogenicity showed a significant overlap, verifying that there are structural commonalities between the two phenomena. CASE also identified biophores significantly associated with the activity of non-genotoxic carcinogens, thereby suggesting the unexpected possibility that there is a structural commonality among the chemicals included in this group. A comparison between the biophores responsible for carcinogenicity in mice and rats resulted in the identification of common ("universal") biophores. It is suggested that agents which contain "universal" biophores are more likely to present a risk to human than carcinogens that do not possess such biophores. CASE also permitted the recognition of species-specific carcinogenic biophores. While the former are primarily electrophiles or potential electrophiles, the latter represent non-electrophilic structures.

Animals↗

PubMed1992

A multipurpose teaching workstation using expert systems, CD ROM and interactive laserdisc.

This demonstration shows a multipurpose workstation used in a clinical teaching application which combines currently available software suitable for clinical diagnosis and teaching. The medical software includes QMR, Scientific American Medicine, the Slice of Life and generic video laserdisc authoring software developed at the University of Ottawa. The system allows a clinical instructor either in an individual or in a small group teaching setting on a ward or in a classroom, to access high quality differential diagnosis information via QMR, which is then supplemented by the text components of Scientific American Medicine on CD ROM, with video laserdisc of the appropriate anatomy, imagery and pathology provided by one of the various laserdiscs. The generic authoring software allows the instructor or students to construct subject related tutorial or testing modules either with or without video laserdisc support. The workstation uses DESQview as the multitasking environment to control the various resources. This program allows easy transfer from one application to another and allows for marking and pasting of text material into a study document. DESQview can also be used to script a specific learning sequence. The demonstration will show the interaction required to study a specific clinical problem and how this can be made into a meaningful multimedia experience with hardcopy for study purposes.

CD-ROM↗

PubMed1996

On the interpretation of certainty factors in expert systems.

Despite the strong theoretical foundation the Bayesian probabilistic approach to model uncertainty in medicine meets many difficulties at the implementation step. One of these difficulties is related to a large amount of conditional probabilities to be assessed and in many cases this task was recognised to be practically insoluble. The MYCIN certainty factors model is a widely distributed pragmatical approach for modeling reasoning under uncertainty that substantially simplifies the problem, at the sacrifice of theoretical soundness. One can determine certainty factors as a function of prior and posterior probability. However, this approach is only consistent with the modularity axiom for certainty factors for tree-structure inference networks, which is rarely true for practical applications. In this paper we abandon the requirement of a direct probabilistic interpretation of certainty factors and build a model of propagation of uncertainty in terms of absolute belief and belief updates. We describe our model for propagating uncertainty in terms of matrix multiplication with specifically defined addition and multiplication which correspond to parallel and sequential combinations of certainty factors. It is possible to define these operations in such a manner that they form a field, and therefore to obtain some useful properties. Finally we present a method of determining certainty factors from statistical data using nonlinear regression and illustrate it with a leukemia diagnostics problem.

Artificial Intelligence↗

PubMed2000

Development of an expert system rulebase for the prospective identification of photoallergens.

Relationships between the structure and properties of chemicals can be programmed into knowledge-based systems such as DEREK (an acronym for 'Deductive Estimation of Risk from Existing Knowledge'). The DEREK knowledge-based computer system contains a sub-set of over 50 rules describing chemical substructures (toxophores) responsible for skin sensitization. This rulebase, based originally on Unilever historical in-house guinea pig maximisation test data, is largely complete and is undergoing refinement as the next stage of its development. As part of an ongoing program of validation and testing, the predictive ability of the sensitization rule set was assessed by processing the structures of over 100 chemical substances in the list of contact allergens identified by the BgVV (German Federal Institute for Health Protection of Consumers). The exercise highlighted areas of chemistry where further development of the rulebase was required, either by extension of the scope of existing rules or by generation of new rules where a sound mechanistic rationale for the biological activity could be established. Several chemicals likely to be acting as photoallergens were identified and rules for photoallergenicity were written covering three classes of chemicals. This paper describes work to extend the DEREK rules for photoallergenicity as part of the European Phototox Project.

Allergens↗

PubMed1993

Graphical access to medical expert systems: V. Integration with continuous-speech recognition.

This paper describes three prototypes of computer-based clinical record-keeping tools that use a combination of window-based graphics and continuous speech in their user interfaces. Although many of today's commercial speech-recognition products achieve high rates of accuracy for large grammars (vocabularies of words or collections of sentences and phrases), they can only "listen for" (and therefore recognize) a limited number of words or phrases at a time. When a speech application requires a grammar whose size exceeds a speech-recognition product's limits, the application designer must partition the large grammar into several smaller ones and develop control mechanisms that permit users to select the grammar that contains the words or phrases they wish to utter. Furthermore, the user interfaces they design must provide feedback mechanisms that show users the scope of the selected grammars. The three prototypes described were designed to explore the use of window-based graphics as control and feedback mechanisms for continuous-speech recognition in medical applications. Our experiments indicate that window-based graphics can be effectively used to provide control and feedback for certain classes of speech applications, but they suggest that the techniques we describe will not suffice for applications whose grammars are very complex.

Decision Making, Computer-Assisted↗

PubMed1990

Therapeutic antibiotic monitoring: surveillance using a computerized expert system.

STUDY OBJECTIVE: To develop and evaluate a computerized system to monitor therapeutic antibiotics in a hospital setting. MATERIAL AND METHODS: From November 1986 through October 1987, we prospectively monitored 1,632 hospitalized patients who had 2,157 microbiology specimens sent for culture and sensitivity testing. During the study period, computer algorithms were used to identify patients whose antibiotic therapy was inappropriate in relation to microbiology culture and sensitivity data. When inconsistencies occurred between antibiotic therapy and in vitro sensitivity data, computer algorithms generated therapeutic antibiotic monitor (TAM) alerts. A clinical pharmacist then notified the attending physician of the alert. RESULTS: Antibiotic therapy was identified by the computer as inappropriate in 696 instances (32%). After we eliminated false-positive alerts, 420 evaluable TAM alerts remained. Physicians responded to the TAM alerts by either changing or starting antimicrobial therapy in 125 cases (30%). Moreover, physicians were previously unaware of the relevant susceptibility test results in 49% of the alerts. CONCLUSION: Computer-assisted monitoring is an efficient and promising method to identify and correct errors in antimicrobial prescribing and to assure the appropriate use of therapeutic antibiotics.

Algorithms↗

PubMed1995

Automated patient care documentation: what's in it for us? An expert system emergency drug card printout.

This article summarizes experiences to date for a collaborative approach, utilizing systems analysis and decision support methods, to design, develop, and implement automated patient centered documentation. Current manual methods for retrieving patient centered data for treatment activities and evaluation of practice are laborious, frustrating and often uneventful. Accessing specific patient information in an arrest or emergent situation, in the hospital and out patient clinical setting, is fraught with difficulties of data and information availability, reliability, legibility, integrity, security, and obsolescence. Treatment decisions made during an arrest or emergent situation, whether for an in-patient or clinical out patient, utilize Advanced Life Support guidelines and also may vary based on the heuristics of the lead practitioner on duty at the time. Walking the informatics talk of "managing and processing data to information to knowledge" lead to standardization for best practice of emergency drug calculations and treatments (1). An expedient and reliable method for retrieving patient specific data to calculate 26 medications, 3 treatments, and upwards of 40 criteria to consider during an arrest or emergent situation was achieved and implemented, as a by product of height and weight charting, across most all patient care areas at Primary Children's Medical Center in Salt Lake City Utah.

Attitude to Computers↗

PubMed1988

A performance evaluation of the expert system ANEMIA.

This paper reports the results of an evaluation study of the current level of performance given by ANEMIA, a knowledge-based consultation system addressing the clinical problem of managing anemic patients. ANEMIA was developed on a mainframe using the AI programming scheme EXPERT and then translated into a version running on a personal computer. At present the system is able to provide assistance in the diagnosis and management of 65 disease entities. After extensive local testing of accuracy, completeness, and consistency of the knowledge base included into ANEMIA, we designed a study to evaluate whether the system is able to appropriately mirror also the reasoning of well-known hematologists other than those who provided the knowledge. We were also interested in testing whether there were conflicting opinions among hematologists. Thus, we designed a validation study in which ANEMIA's performance could be compared with that of six hematologists and the interexpert consensus evaluated. ANEMIA's overall performance was judged acceptable in 87% (26/30) of the cases, while expert evaluators agreed with their colleagues in 90% (27/30) of them. A low interexpert consensus was found: considering the ratings given by different hematologists to the same ANEMIA performance, complete agreement occurred only 47% of the time.

Adult↗

PubMedDate not supplied

Variables processing in expert system building: application to the aetiological diagnosis of infantile meningitis.

We describe a method for the analysis and quantitative synthesis of information in a clinical data file using as an example the aetiological diagnosis of infantile meningitis. The results obtained together with the qualitative knowledge of experts in the field have enabled an experimental system to be constructed. This is performing sufficiently well to be introduced into a hospital environment to assist young doctors facing an emergency situation.

Acute Disease↗

PubMedDate not supplied

[An expert system for biopsy diagnosis of lymph node pathology using a personal computer].

The program has been developed for biopsy diagnosis in lymph node disease including tumors. The expert knowledge is based on the information about more than 200 morphological symptoms of 37 nosological unities: reactive lymphoid hyperplasia, metastatic tumors, granulomatous disease and others. The decision three consists of three sections: diffuse nontumor processes, inflammatory focal processes and tumor diffuse and focal processes. The system can provide a "second opinion" for pathologists in their professional activity and may useful as a teaching tool for medical students.

Biopsy↗

PubMed1987

[Current computerized support for vestibular function tests. II. An expert system for the classification of vertigo].

A computerized system (VERTIGO) aimed to the classification and diagnosis of different types of vertigo has been developed. It is based on the shell EXPERT (Weiss and Kulikowski, 1979). At present only the findings arising from patient history are considered as input data and the diagnostic possibilities of the system have been limited to the differential diagnosis of vertigo due to peripheral vestibular disorders. About thirty different forms of vertigo are taken into account. They are clustered in two groups, true vertigo and dizziness. Data are collected through a computer controlled questionnaire. Using the facilities offered by EXPERT, the sequence of the questions can be modified according to the flow of information in order to reproduce different diagnostic strategies. After history taking, the system presents a summary of the available findings followed by its diagnostic conclusions. Different conclusions can be proposed with different degrees of certainty. Conclusions can be justified by the system when required.

Adult↗

PubMed1990

Hypermedia and randomized algorithms for medical expert systems.

KNET is an environment for constructing probabilistic, knowledge-intensive systems within the axiomatic framework of decision theory. The KNET architecture defines a complete separation between the hypermedia user interface on the one hand, and the representation and management of expert opinion on the other. KNET offers a choice of algorithms for probabilistic inference. We and our coworkers have used KNET to build consultation systems for lymph-node pathology, bone-marrow transplantation therapy, clinical epidemiology, and alarm management in the intensive-care unit. Most important, KNET contains a randomized approximation scheme (RAS) for the difficult and almost certainly intractable problem of Bayesian inference. Our algorithm can, in many circumstances, perform efficient approximate inference in large and richly interconnected models of medical diagnosis. In this article, we describe the architecture of KNET, construct a randomized algorithm for probabilistic inference, and analyze the algorithm's performance. Finally, we characterize our algorithms' empiric behavior and explore its potential for parallel speedups. From design to implementation, then, KNET demonstrates the crucial interaction between theoretical computer science and medical informatics.

Algorithms↗

PubMed1998

[The use of computer-assisted clinical consultation with expert systems in early pregnancy].

OBJECTIVES: The early pregnancy disorders are most frequent diagnostic problem in a GP-practice. This observation and commonness of the problem inclined the authors to assume eutopic physiological pregnancy, eutopic complicated pregnancy, ectopic pregnancy and trophoblastic diseases as important diagnostic hypotheses for our knowledge based consultation system, intended as decision support tool for first contact physician (GP) when diagnosing early pregnancy disorders. DESIGN: Defining 81 diagnostic tests to be included in the knowledge acquisition process, interviewing clinical experts and developing knowledge base (set of clinical diagnostic rules), implementing the knowledge base for early pregnancy disorders--related amenorrhea patients, in particular: determining clinical importance of symptoms and lab/diagnostic tests. MATERIALS AND METHODS: The clinical knowledge acquisition process has been performed taking into account 81 diagnostic features. Moreover clinical data on 220 cases, including 96 early pregnancy disorders cases, have been collected to enable validation of the knowledge base. ELSA approach to creating environment-self-adjustable medical knowledge bases, developed by Ruszkowski has been applied and the eleven main diagnostic hypotheses have been defined. We believe that the proposed knowledge base and PC-based consultation system offers such an approach for improving GP diagnostic effectiveness. RESULTS AND CONCLUSIONS: 1. The knowledge base and the consultation system could contribute to improving diagnostic effectiveness, when used by a GP for diagnosing early pregnancy disorders such an approach could also be used to determine relative diagnostic power of specific tests or symptoms, 2. The laboratory and diagnostic test confirmed high diagnostic accuracy in early pregnancy disorders. The anamnesis has a low diagnostic effectiveness for particular diagnostic features in ectopic pregnancy patients and physical examination confirmed low diagnostic accuracy in eutopic physiological pregnancy and eutopic complicated pregnancy.

Adult↗

PubMed1997

Prototype expert system to assist with the stabilization of neonates prior to transport.

The transport of sick or premature newborns from community hospitals to acute care facilities is often necessary for the infants to receive the level of care required. Prior stabilization of these infants is imperative to a successful transport. The knowledge required to treat and stabilize sick and premature newborns is specialized and may not be available in community hospitals. "The S.T.A.B.L.E. Assistant" is a prototype rule based decision support system based on the educational program "Transporting Newborns the S.T.A.B.L.E. Way". "The S.T.A.B.L.E. Assistant" accepts clinical information related to an infant's breathing status, blood glucose status and selected lab values and provides instructions as to what interventions are needed to stabilize and prepare the infant for transport. The computerized program was evaluated using data collected from 19 charts of newborns requiring transport from a community hospital to a tertiary care center. Patient data were entered into the computerized program and the subsequent instructions were reviewed by a group of neonatology and neonatal transport experts. Reviewers evaluated "The S.T.A.B.L.E. Assistant" for the safety of the program, the extent to which the computerized program follows the educational program guidelines and the degree to which the instructions generated are within the scope of community caregivers' practice. Results were positive, indicating "The S.T.A.B.L.E. Assistant" prototype is safe, and within the guidelines of neonatal clinical practice.

Decision Support Systems, Clinical↗

PubMedDate not supplied

High performance for expert systems: II. A system for leukaemia diagnosis.

A system for diagnosing leukaemia subtypes from monoclonal antibody test results has been programmed directly from the expertise of one immunologist. It gives an acceptable conclusion for 400 past cases, with summaries of its reasoning and suggestions for further tests. Its development included three major components: elicitation of the basic items of knowledge, formulation of a coordinated overall scheme, and representation in a computer language. All three were interdependent and required many cycles of trial and error, involving further discussions with the expert at all stages. By far the most difficult component was the formulation of a comprehensive specification for the problems of the domain, and this was only achieved when the knowledge engineer had gained a fair understanding of the application. The system is a tree-structured logic program written in PROLOG. This format facilitated experimentation with different knowledge representations, control of the interrelationships between rules and the elimination of errors. The problems of quality control and maintenance are of paramount importance for high-performance systems.

Diagnosis, Computer-Assisted↗

PubMedDate not supplied

Development of decision making rules for transportable, microcomputer-based expert systems in medicine.

Computerized decision making aids have been developed for use in certain medical specialties over the last decade. These programs are designed to give expert-level advice within their limited domains. Two major components are required in the design of these systems: development of software and development of a knowledge base to handle the specific medical application. The advantage of the knowledge-based approach is that the application can be changed by replacement of the knowledge base, without requiring changes to the software. In this paper, EMERGE, a medical decision making system, originally developed for analysis of chest pain in the emergency room environment, is described. EMERGE uses production rules for representation of its knowledge base. The method by which this system can accommodate new rule bases for other applications is emphasized, as well as the transportability of the system.

Computers↗