SEARCH · PubMed Health
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.
This paper describes extensions to the Mumps language which facilitate writing expert systems in Mumps. These extensions are based upon the Prolog language and utilize the existing Mumps global array data base as well as other aspects of the Mumps environment. Details are given of an experimental implementation of the interpreter written in C for the Unix operating system.
Focus is an expert system developed to diagnose, by the findings on neurological examination, the anatomic location of an acute neurologic event. The inference engine uses a forward tracking algorithm and a rule based data base that includes associations of neurological signs with specific anatomic areas. Focus is linked to a data base in which all personal cases and final diagnoses confirmed by laboratory studies are stored. This makes it possible to assess the validity of the system's diagnoses and use personal cases as additional example to the revision of the rules. The program is easy to use and runs on MS-DOS microcomputer. Furthermore the diagnostic rules are accessible so Focus may furnish an interactive teaching device.
Making a systematic and comprehensive psychiatric evaluation of mental disorders in a patient can be a rather complex and involving process. We describe an expert system, MILP, which is designed to produce such systematic diagnoses of mental disorders using selected categories from the classification and diagnostic guidelines published in DSM-III-R, DSM-IV and ICD-10. An innovative part of the MILP design is the incorporation of constraint-based reasoning as a key part of the system. We believe that the MILR design gives a flexible framework which is suitable in general for the automated diagnoses of large classes of mental disorders.
Quantitative evaluation of upper extremity impairment, a percentage rating most often determined using a rule based procedure, has been implemented on a personal computer using an artificial intelligence, rule-based expert system (AI system). In this study, the rules given in Chapter 3 of the AMA Guides to the Evaluation of Permanent Impairment (Third Edition) were used to develop such an AI system for the Apple Macintosh. The program applies the rules from the Guides in a consistent and systematic fashion. It is faster and less error-prone than the manual method, and the results have a higher degree of precision, since intermediate values are not truncated.
To aid in the validation and interpretive reporting of results from multianalyte diagnostic testing for pheochromocytoma (PHEO), we have developed a desktop personal computer-based laboratory expert system (REPCAT). REPCAT utilizes a commercial database application to run procedures that assess analytical and clinical data relating to patient urine or plasma samples. REPCAT was used to evaluate the raw data from >4000 24-h urine samples submitted to our laboratory for testing for the presence of PHEO. REPCAT performed equivalently to an expert pathologist in assessing the presence and class of PHEO (epinephrine, norepinephrine, or mixed secretor). No false negatives were generated and it assigned a correct interpretation (on the basis of subsequent clinical and biochemical investigation) for each of the primary diagnostic samples from these patients.
The validation of a clinical laboratory report is a process that guarantees the results contained in the report have been obtained under satisfactory metrological conditions and that they are compatible with the information available on the patient. This validation is generally carried out manually by a clinical laboratory professional, but also may be done by an expert system properly programmed, such as the VALAB system. The evaluation presented in this article consists of comparing human and system decisions of validation for 500 randomly selected clinical laboratory reports from hospitalized patients. In this evaluation, 84.8% of the reports examined by the VALAB are accepted directly without any human aid, and only 15.2% require examination by clinical biochemists.
PURPOSE: To clarify terms surrounding the clinical problem of initiating oral feeding in premature infants cared for in the neonatal intensive care unit (NICU). Understanding terminology at the outset of expert system (ES) development forms the crucial foundation needed to build a system that is accurate, relevant, and useful for assisting clinicians. METHOD: Ethnographic methods were used. In phase 1 participant observation with an interdisciplinary neonatology team was used to study oral feeding for premature infants. In phase 2 in-depth interviews were conducted with a neonatal nurse practitioner (NNP) to explore feeding terms and the description and understanding of these terms. FINDINGS: Results showed that oral feeding was a complex term that encompassed several feeding options; specifically nipple, breast, gavage, and transpyloric feeding. CONCLUSIONS: Each of the terms had a specific meaning within the context of the NICU.
Computers have become commonplace in nearly every setting, from home to office. In addition to their usefulness for managerial tasks, they have the potential to aid the primary decision-making tasks of professional social service workers. The benefits and background of such "expert systems" are explored, and the structure of one experimental consulting program for child welfare workers, PLACECON, is presented.
In the field of medicine, ever greater use is being made of modern data processing developments. The paramount targets are quality assurance and/or improvement of patient care. For the diagnosis of renal diseases and their courses of therapy, interesting knowledge-based expert systems are now available that make an important contribution to optimizing urine findings and detecting proteinuria while the course of disease is being controlled.
Three-dimensional reconstruction of serial sections, observed by microscopy, is an important technique in medicine and biology. To view any part of a structure clearly, that part has to be identified and clearly highlighted in its relationship to other features in a structure. The identification process can be time consuming and tedious if many sections are involved, especially for routine applications, since human identification is required. In this paper we describe how image processing, together with other information on the shape and position of features relative to each other on any one section and throughout the structure, could be incorporated into an expert system and we also show how such a system could be designed. An important feature is the use of human-computer interaction to allow the system to evolve under the guidance of the biological or medical expert. An example of feature identification in a plant-parasitic nematode is used.
Based on the quantitative determination of creatinine, total protein, albumin, alpha 1-microglobulin, IgG, alpha 2-macroglobulin, and N-acetyl-beta, D-glucosaminidase in urine in combination with a test strip screening, the findings of hematuria, leukocyturia, and proteinuria can be assigned to prerenal, renal, or postrenal causes. Using this graded diagnostic strategy as a knowledge base, we developed a computerbased expert system for urine protein differentiation ("UPES") as a decision-supporting tool. The knowledge base was implemented as a combination of "if/then" rules and two-step bivariate distance classification of marker proteins. The knowledge for this form of pattern recognition was derived from the results for a set of 267 patients with clinically and histologically documented nephropathies. To determine the diagnostic value of UPES, we tested another set of data: results for 129 urine analyses from 94 patients. Using these data, the system reached 98% concordance with the clinical diagnoses for the patients and was superior to the diagnostic interpretations of four human experts. UPES has been successfully integrated into the laboratory routine process, including automated data import.
Conceptual models for diagnostic reasoning proposed in the medical literature are presented to stimulate discussion about the issue of the appropriateness of probabilistic knowledge-based systems for medical diagnosis. Evidence is presented to corroborate the authors' view that diagnosis is a problem-solving task, rather than a decision-making task. In the authors' opinion, probabilistic reasoning is better suited to situations dealing with choices for clinical intervention, rather than to those dealing with determining the correct diagnosis. A critique is given of a diagnostic Bayesian expert system for lymph node pathology. In empirical studies, diagnostic Bayesian systems have been shown to typically list the correct diagnosis as the program's first choice 60% to 70% of the time. One reason for this undistinguished level of diagnostic performance is that Bayesian systems are not designed to represent and use knowledge the same way that an expert does.
OBJECTIVE: Fluorodeoxyglucose (FDG) positron emission tomography (PET) is increasingly used to support a diagnosis of Alzheimer's disease. The aim of the present study was to evaluate a new expert system (PALZ) for the fully automated analysis of FDG PET images for diagnosis of the disease. METHODS: The PALZ tool is based on the detection of the typical disease pattern in FDG PET images. Its potential for this task was evaluated in 22 consecutive patients with suspected Alzheimer's disease who had been graded as positive for the pattern by an experienced reader (visual analysis supported by statistical parametric mapping (SPM)), and in 18 controls. Dependence on scanner performance was assessed by variation of the spatial resolution of the PET images. RESULTS: All the Alzheimer's disease subjects were classified as pattern-positive by the PALZ tool. Fifteen controls were classified as normal. Sensitivity and specificity for differentiation of the patients with suspected Alzheimer's disease from the controls were 100% and 83%, respectively. The false positive finding in three controls most likely was caused by differences in attenuation correction between the normal data base of the PALZ tool (cold transmission scan) and the local data sets (hot transmission scan). There was only mild dependence on spatial resolution. CONCLUSIONS: The results of the present study suggest that the PALZ tool provides similar performance for the detection of the typical Alzheimer's disease pattern in FDG PET images as an experienced reader supported by SPM. The PALZ tool is fully automated, easy to use, and insensitive to the spatial resolution of the PET scanner used. Therefore, it has the potential for widespread clinical use.
Experience with an expert system developed for the purpose of classification of anemias is presented. Input for this system consists of limited demographic information on each patient and the results of the complete blood count, with the incorporation of the results of further chemical testing (serum iron/total iron binding capacity/ferritin and serum B12/serum folate/red blood cell folate), if indicated. Performance of this system is demonstrated using data from 84 patients seen in the authors' laboratory selected either because of a request for evaluation by the attending physician or because of significant anemias. Using this limited input, the system was able to accurately classify 74 of 84 (88%) of cases according to previously established criteria. The output from the system is overread by a pathologist. Future directions are also discussed.
The Advanced Expert System (AES) was used in conjunction with the VITEK 2 automated antimicrobial susceptibility test system to ascertain the beta-lactam phenotypes of 196 isolates of the family Enterobacteriaceae and the species Pseudomonas aeruginosa. These isolates represented a panel of strains that had been collected from laboratories worldwide and whose beta-lactam phenotypes had been characterized by biochemical and molecular techniques. The antimicrobial susceptibility of each isolate was determined with the VITEK 2 instrument, and the results were analyzed with the AES to ascertain the beta-lactam phenotype. The results were then compared to the beta-lactam resistance mechanism determined by biochemical and molecular techniques. Overall, the AES was able to ascertain a beta-lactam phenotype for 183 of the 196 (93.4%) isolates tested. For 111 of these 183 (60.7%) isolates, the correct beta-lactam phenotype was identified definitively in a single choice by the AES, while for an additional 46 isolates (25.1%), the AES identified the correct beta-lactam phenotype provisionally within two or more choices. For the remaining 26 isolates (14.2%), the beta-lactam phenotype identified by the AES was incorrect. However, for a number of these isolates, the error was due to remediable problems. These results suggest that the AES is capable of accurate identification of the beta-lactam phenotypes of gram-negative isolates and that certain modifications can improve its performance even further.
A subpackage for spectral interference correction has been set up in ICP-AES primary expert system. Some numerical recipes have been applied in this subpackage, such as curve smoothing, derivative spectrum, FFT technique, spectral resolving etc. A comprehensive method for spectral interference correction is presented. All these numeric recipes have been tested with simulated data and also utilized in several typical examples of real spectra. Results are satisfactory.
Computerized information systems that support medical decision-making are based upon formalized medical knowledge. Therefore, appropriate methods of knowledge engineering are of conspicuous importance. In order to arrive at a medical-history expert system, a target-oriented construction method of knowledge engineering was devised. Beginning with a framework of diagnostic categories, an optimal set of reliable and valid questions were formulated. In this paper, two independently developed general medicine questionnaires prepared by such a method are comparatively presented. It is shown that (a) inspite of a good deal of similarity between the two frames of reference considerable differences exist between the two sets of selected questions, and (b) the valid collection of medical information is limited to diagnostic categories that allow a clear description of medical symptoms.
Our ongoing efforts in health services research have resulted in outcome-validated medical practice guidelines for common medical conditions. These practice guidelines have been shown to substantially reduce health care costs while maintaining quality of care. We have developed a computerized expert system from our practice guidelines which enhances the ease in which they can be implemented by Utilization Management (UM) Coordinators, physicians, nurses and other health care providers.