Search PubMed⌕ Search

SEARCH · Search PubMed

Results for “Intelligent Systems”

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 919 records · Page 51Linked to original sources
PubMed2001

[Progress in knowledge-based X-ray fluorescence spectrometry].

The review focuses on the expert systems and knowledge engineering in X-ray fluorescence spectrometry. It includes mainly a knowledge-controlled strategy combining the scan-based method with the fixed channel measurements, a XRF interpretation system of spectra with fuzzy logic and pattern recognition, and an expert system for qualitative interpretation of XRF spectra using a certainty factor. In the review, a series of the studies of exploring the knowledge engineering system in XRF are also included, which consists of four parts, i.e. spectra identification, pattern recognition with decision-making, quantitative determination combined with the theoretical alpha coefficients and neural networks, and XRF analysis without standards.

Algorithms↗
PubMedDate not supplied

Learning from biomedical time series through the integration of qualitative models and fuzzy systems.

Our work deals with a method for the identification of the dynamics of nonlinear (patho-)physiological systems by learning from data. The key idea which underlies our approach consists in the integration of qualitative modeling methods with fuzzy logic systems. The major advantage which derives from such an integrated framework lies in its capability both to represent the structural knowledge of the system at study and to determine, by exploiting the available experimental data, a functional approximation of the system dynamics that can be used as a reasonable predictor of the patient's future state. We have successfully applied our method in the identification of the intracellular kinetics of thiamine from data collected in the intestine cells.

Artificial Intelligence↗
PubMed2026

AI-driven snapshot hyperspectral imaging for on-line sorting systems in food industry: From real-time sensing to intelligent decision-making.

High-throughput food sorting requires rapid, non-destructive detection of external defects, foreign materials, and internal quality attributes in heterogeneous food matrices. Conventional scanning hyperspectral imaging may suffer from motion-induced spatial-spectral mismatches, whereas snapshot hyperspectral imaging (S-HSI) captures spectral images within a single integration time. However, its advantage is limited by trade-offs in resolution, signal-to-noise ratio (SNR), reconstruction uncertainty, and calibration stability, which are further amplified by variable tissue structure, surface reflection, moisture, and fat distribution in foods. This review critically examines artificial intelligence (AI)-driven S-HSI for on-line food sorting within a sensing-representation-decision-execution framework. Compact architectures are compared according to their physical constraints, food-sorting suitability, and ability to support mapping between spectral responses and physicochemical quality attributes. AI strategies are reviewed for spectral reconstruction, image restoration, spatial-spectral representation, band selection, uncertainty-aware decision-making, and edge implementation. AI can partially compensate for snapshot-specific limitations, but current evidence remains largely limited to laboratory or prototype studies. Future work should link system performance to food safety and quality outcomes by reporting throughput, decision latency, calibration drift, missed-detection risk, false-rejection cost, and closed-loop sorting success.

Hyperspectral Imaging↗
PubMed1997

Computational predictive programs (expert systems) in toxicology.

The increasing number of pollutants in the environment raises the problem of the toxicological risk evaluation of these chemicals. Several so called expert systems (ES) have been claimed to be able to predict toxicity of certain chemical structures. Different approaches are currently used for these ES, based on explicit rules derived from the knowledge of human experts that compiled lists of toxic moieties for instance in the case of programs called HazardExpert and DEREK or relying on statistical approaches, as in the CASE and TOPKAT programs. Here we describe and compare these and other intelligent computer programs because of their utility in obtaining at least a first rough indication of the potential toxic activity of chemicals.

Animals↗
PubMed1994

Speech intelligibility and protective effectiveness of selected active noise reduction and conventional communications headsets.

An experiment was conducted to compare both speech intelligibility and noise attenuation of a conventional passive headset (David Clark H10-76) and an electronic Active Noise Reduction (ANR) headset (Bose Aviation) operated with and without its ANR feature. Modified Rhyme Tests were conducted in pink and tank noise, and with and without bilateral phase reversal between earphones. The Bose ANR unit required a significantly higher speech-to-noise (S/N) ratio in both noise environments than the two passive headset systems to maintain equal intelligibility, in part because of its stronger noise reduction and higher required signal level. Articulation Index calculations corroborated the empirical result that the David Clark afforded comparable intelligibility to the Bose ANR device. Bilateral phase reversal proved to be of no benefit, and pink noise proved to be the harsher environment for speech intelligibility. On a speech intelligibility basis alone, the results do not justify the additional cost of the ANR headset; however, when severe noise exposure is at issue, a properly functioning ANR unit may afford more protection than a similar passive headset without electronics, especially in low-frequency noise spectra.

Acoustics↗
PubMed1996

Architectural design and tools to support the transparent access to hospital information systems, radiology information systems, and picture archiving and communication systems.

The fragmentation of the electronic patient record among hospital information systems (HIS), radiology information systems (RIS), and picture archiving and communication systems (PACS) makes the viewing of the complete medical patient record inconvenient. The purpose of this report is to describe the system architecture, development tools, and implementation issues related to providing transparent access to HIS, RIS, and PACS information. A client-mediator-server architecture was implemented to facilitate the gathering and visualization of electronic medical records from these independent heterogeneous information systems. The architecture features intelligent data access agents, run-time determination of data access strategies, and an active patient cache. The development and management of the agents were facilitated by data integration CASE (computer-assisted software engineering) tools. HIS, RIS, and PACS data access and translation agents were successfully developed. All pathology, radiology, medical, laboratory, admissions, and radiology reports for a patient are available for review from a single integrated workstation interface. A data caching system provides fast access to active patient data. New network architectures are evolving that support the integration of heterogeneous software subsystems. Commercial tools are available to assist in the integration procedure.

Computer Systems↗
PubMed2001

Achieving reuse of computable guideline systems.

We describe an architecture for reusing computable guidelines and the programs used to interpret them across varied legacy clinical systems. Developed for the PRODIGY 3 project, our architecture aims to support interactive, point of care use of guidelines in primary care. Legacy medical record systems in UK primary care are diverse, using different terminologies, different data models, and varying user-interface philosophies. However, our goal is to provide common guideline knowledge bases and system components, while achieving full integration with the host medical record system, and a user interface tailored to that system. In conjunction with system suppliers, we identified areas of standardization required to achieve this goal. Firstly, standardized interfaces were created for mediation with the legacy system medical record and for act management. Secondly, a standard interface was developed for communication with the User Interface for guideline interaction. Thirdly, a terminology mapping knowledge base and system component was provided. Lastly, we developed a numeric unit conversion knowledge base and system component. The standardization of this architecture was achieved by close collaboration with existing vendors of Primary Care computing systems in the UK. The work has been verified by two suppliers successfully building and deploying systems with User Interfaces which mirror their normal look and feel, communicating fully with existing medical records, while using identical Guideline Interpreter components and knowledge bases. Encouragingly further experiments in other areas of clinical decision support have not required extension of our interfaces.

Artificial Intelligence↗
PubMed1994

MD Concept: a model for integrating medical knowledge.

Many integrated clinical information systems depend on large knowledge bases containing dictionary of terms as well as specific information about each term and the relationships between terms. We propose a knowledge base model called MD Concept which is based on a semantic network and uses an object-oriented paradigm and relational tables. A prototype has been developed which integrates the Unified Medical Language System (UMLS) with other databases including the Systematized Nomenclature of Medicine (SNOMED II), the Diagnostic and Statistical Manual of Mental Disorders (DSM-IIIR) and a pharmaceutical database. We demonstrate how a user can easily navigate in this knowledge world using a browser.

Artificial Intelligence↗
PubMed1990

Knowledge bases in medicine: a review.

Efforts to represent knowledge effectively have been central to progress in various aspects of medical informatics. These efforts range from relatively simple "electronic textbooks" to fairly sophisticated knowledge-based systems, which function as well as, or even better than, human experts faced with similar problems. Knowledge bases have been developed in many fields, but the relatively limited domains and structured language of medicine, as well as the importance of information in the provision of good medical care, have made research in medical knowledge representation an area of intense activity. This paper reviews representative knowledge bases and knowledge-based systems in medicine: electronic textbooks such as PDQ and the Hepatitis Knowledge Base (HKB), rule-based systems such as MYCIN, causal models (e.g., CASNET), and hypothesis- or frame-based systems, exemplified by PIP and INTERNIST-1. The paper describes the relationships among divergent approaches and provides a sense of current and future trends. It examines problems in knowledge-based systems, particularly in knowledge representation and acquisition, and the responses to these challenges. The latter include the use of domain-independent software shells for constructing knowledge bases, the adaptation and use of previously existing knowledge bases, and multiple uses of the same knowledge base for different purposes.

Artificial Intelligence↗
PubMed2006

Classification of a known sequence of motions and postures from accelerometry data using adapted Gaussian mixture models.

Accelerometry shows promise in providing an inexpensive but effective means of long-term ambulatory monitoring of elderly patients. The accurate classification of everyday movements should allow such a monitoring system to exhibit greater 'intelligence', improving its ability to detect and predict falls by forming a more specific picture of the activities of a person and thereby allowing more accurate tracking of the health parameters associated with those activities. With this in mind, this study aims to develop more robust and effective methods for the classification of postures and motions from data obtained using a single, waist-mounted, triaxial accelerometer; in particular, aiming to improve the flexibility and generality of the monitoring system, making it better able to detect and identify short-duration movements and more adaptable to a specific person or device. Two movement classification methods were investigated: a rule-based Heuristic system and a Gaussian mixture model (GMM)-based system. A novel time-domain feature extraction method is proposed for the GMM system to allow better detection of short-duration movements. A method for adapting the GMMs to compensate for the problem of limited user-specific training data is also proposed and investigated. Classification performance was considered in relation to data gathered in an unsupervised, directed routine conducted in a three-month field trial involving six elderly subjects. The GMM system was found to achieve a mean accuracy of 91.3%, distinguishing between three postures (sitting, standing and lying) and five movements (sit-to-stand, stand-to-sit, lie-to-stand, stand-to-lie and walking), compared to 71.1% achieved by the Heuristic system. The adaptation method was found to offer a mean accuracy of 92.2%; a relative improvement of 20.2% over tests without subject-specific data and 4.5% over tests using only a limited amount of subject-specific data. While limited to a restricted subset of possible motions and postures, these results provide a significant step in the search for a more robust and accurate ambulatory classification system.

Acceleration↗
PubMed2003

Effects of PCB exposure on neuropsychological function in children.

In the last decade advances in the analytic methods for quantification of polychlorinated biphenyls (PCBs) have resulted in widespread availability of congener-specific analysis procedures, and large amounts of data on PCB congener profiles in soil, air, water, sediments, foodstuffs, and human tissues have become available. These data have revealed that the PCB residues in environmental media and human tissues may not closely resemble any of the commercial PCB mixtures, depending on source of exposure, bioaccumulation through the food chain, and weathering of PCBs in the environment. At the same time, toxicological research has led to a growing awareness that different classes of PCB congeners have different profiles of toxicity. These advances in analytic techniques and toxicological knowledge are beginning to influence the risk assessment process. As the data from ongoing PCB studies assessing the mediators of neurobehavioral outcomes in children are published, the weight of evidence for PCB effects on neurodevelopment is growing. Studies in Taiwan, Michigan (USA), New York (USA), Holland, Germany, and the Faroe Islands have all reported negative associations between prenatal PCB exposure and measures of cognitive functioning in infancy or childhood. The German study also reported a negative association between postnatal PCB exposure and cognitive function in early childhood--a result that had not been found in previous studies. Only one published study in North Carolina (USA) has failed to find an association between PCB exposure and cognitive outcomes. Despite the fact that several more recent studies have used congener-specific analytic techniques, there have been only limited attempts to assess the role of specific PCB congeners or classes of congeners in mediating neurodevelopmental outcomes. From a statistical standpoint, attempts to determine the role of individual congeners in mediating outcomes are hampered by the fact that concentrations of most individual congeners are highly correlated with each other and with total PCBs. From a toxicological standpoint, these efforts are hampered by the fact that many of the PCB congeners present in human tissues have never been studied in the laboratory, and their relative potency to produce nervous system effects is unknown. More complete information on the health effects of various congeners or congener classes would allow more informed scientific and risk assessment decisions.

Child↗
PubMed2005

Comparative experiments on learning information extractors for proteins and their interactions.

OBJECTIVE: Automatically extracting information from biomedical text holds the promise of easily consolidating large amounts of biological knowledge in computer-accessible form. This strategy is particularly attractive for extracting data relevant to genes of the human genome from the 11 million abstracts in Medline. However, extraction efforts have been frustrated by the lack of conventions for describing human genes and proteins. We have developed and evaluated a variety of learned information extraction systems for identifying human protein names in Medline abstracts and subsequently extracting information on interactions between the proteins. METHODS AND MATERIAL: We used a variety of machine learning methods to automatically develop information extraction systems for extracting information on gene/protein name, function and interactions from Medline abstracts. We present cross-validated results on identifying human proteins and their interactions by training and testing on a set of approximately 1000 manually-annotated Medline abstracts that discuss human genes/proteins. RESULTS: We demonstrate that machine learning approaches using support vector machines and maximum entropy are able to identify human proteins with higher accuracy than several previous approaches. We also demonstrate that various rule induction methods are able to identify protein interactions with higher precision than manually-developed rules. CONCLUSION: Our results show that it is promising to use machine learning to automatically build systems for extracting information from biomedical text. The results also give a broad picture of the relative strengths of a wide variety of methods when tested on a reasonably large human-annotated corpus.

Algorithms↗
PubMed1990

Knowledge-based computer systems for radiotherapy planning.

Radiation therapy is one of the first areas of clinical medicine to utilize computers in support of routine clinical decision making. The role of the computer has evolved from simple dose calculations to elaborate interactive graphic three-dimensional simulations. These simulations can combine external irradiation from megavoltage photons, electrons, and particle beams with interstitial and intracavitary sources. With the flexibility and power of modern radiotherapy equipment and the ability of computer programs that simulate anything the machinery can do, we now face a challenge to utilize this capability to design more effective radiation treatments. How can we manage the increased complexity of sophisticated treatment planning? A promising approach will be to use artificial intelligence techniques to systematize our present knowledge about design of treatment plans, and to provide a framework for developing new treatment strategies. Far from replacing the physician, physicist, or dosimetrist, artificial intelligence-based software tools can assist the treatment planning team in producing more powerful and effective treatment plans. Research in progress using knowledge-based (AI) programming in treatment planning already has indicated the usefulness of such concepts as rule-based reasoning, hierarchical organization of knowledge, and reasoning from prototypes. Problems to be solved include how to handle continuously varying parameters and how to evaluate plans in order to direct improvements.

Artificial Intelligence↗
PubMed1996

Clinical decision support for physician order-entry: design challenges.

We report on a joint development effort between ALLTEL Information Services Health Care Division and IBM Worldwide Healthcare Industry to demonstrate concurrent clinical decision support using Arden Syntax at order-entry time. The goal of the partnership is to build a high performance CDS toolkit that may be easily customized for multiple health care enterprises. Our work uses and promotes open technologies and health care standards while building a generalizable interface to a legacy patient-care system and clinical database. This paper identifies four areas of design challenges and solutions unique to a concurrent order-entry environment: the clinical information model, the currency of the patient virtual chart, the granularity of event triggers and rule evaluation context, and performance.

Artificial Intelligence↗
PubMed1990

Metadata-based generation and management of knowledgebases from molecular biological databases.

Present-day knowledge-based systems (or expert systems) and databases constitute 'islands of computing' with little or no connection to each other. The use of software to provide a communication channel between the two, and to integrate their separate functions, is particularly attractive in certain data-rich domains where there are already pre-existing database systems containing the data required by the relevant knowledge-based system. Our evolving program, GENPRO, provides such a communication channel. The original methodology has been extended to provide interactive Prolog clause input with syntactic and semantic verification. This enables automatic generation of clauses from the source database, together with complete management of subsequent interfacing to the specified knowledge-based system. The particular data-rich domain used in this paper is protein structure, where processes which require reasoning (modelled by knowledge-based systems), such as the inference of protein topology, protein model-building and protein structure prediction, often require large amounts of raw data (i.e., facts about particular proteins) in the form of logic programming ground clauses. These are generated in the proper format by use of the concept of metadata.

Artificial Intelligence↗
PubMedDate not supplied

Insulin pump therapy in type 1 diabetes mellitus.

OBJECTIVES: To review the current experience with insulin pump therapy in children and adolescents in order to guide pediatricians regarding indications and complications. SOURCES OF DATA: Systematic review of articles published in the literature referring to the use of insulin pump therapy, indications, complications and response to treatment. All articles published between 1995 and 2005 and appearing in the MEDLINE and LILACS databases were reviewed. The keywords were: insulin pump, type 1 diabetes mellitus and diabetes mellitus. The articles covering the subject of interest and referring to children and adolescents were selected. SUMMARY OF THE FINDINGS: Insulin pump therapy is not required for all patients with type 1 diabetes, since intensive treatments produce very similar results in terms of glycated hemoglobin and control of complications over the medium and long terms. However, the pump allows for greater comfort for patients, with less rigid meal schedules and better quality of life. The first requirement for patients intending to use the pump is getting used to having a device attached to the body and following strict glucose control; otherwise, pump therapy is not advantageous. Complications are rare due to the technologies currently available. The cost, however, is greater than with conventional treatments. CONCLUSION: The development of infusion pumps and glucose monitors, including continuous monitoring systems, will lead to "intelligent pumps," so that a true "artificial pancreas" will be available, which can even be implanted in the patient, allowing non-diabetic persons to lead a normal life.

Adolescent↗
PubMed1995

Neural network based dynamic controllers for industrial robots.

The industrial robot's dynamic performance is frequently measured by positioning accuracy at high speeds and a good dynamic controller is essential that can accurately compute robot dynamics at a servo rate high enough to ensure system stability. A real-time dynamic controller for an industrial robot is developed here using neural networks. First, an efficient time-selectable hidden layer architecture has been developed based on system dynamics localized in time, which lends itself to real-time learning and control along with enhanced mapping accuracy. Second, the neural network architecture has also been specially tuned to accommodate servo dynamics. This not only facilitates the system design through reduced sensing requirements for the controller but also enhances the control performance over the control architecture neglecting servo dynamics. Experimental results demonstrate the controller's excellent learning and control performances compared with a conventional controller and thus has good potential for practical use in industrial robots.

Algorithms↗