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 1,657 records · Page 92Linked to original sources
PubMed2004

Intelligent motion control for linear piezoelectric ceramic motor drive.

Since the dynamic characteristics of a linear piezoelectric ceramic motor (LPCM) are highly nonlinear and time varying, it is difficult to design a suitable motor drive and position controller that realizes accurate position control at all time. This study investigates a double-inductance double-capacitance (LLCC) resonant driving circuit and a sliding-mode fuzzy-neural-network control (SMFNNC) system for the motion control of an LPCM. First, the motor structure and LLCC driving circuit of an LPCM are introduced. The LLCC resonant inverter is designed to operate at an optimal switching frequency such that the output voltage will not be influenced by the variation of quality factor. Moreover, a SMFNNC system is designed to achieve favorable tracking performance without precise dynamic models being controlled. All adaptive learning algorithms in the SMFNNC system are derived in the sense of Lyapunov stability analysis, so that system-tracking stability can be guaranteed in the closed-loop system. The effectiveness of the proposed driving circuit and control system is verified by experimental results.

Artificial Intelligence↗
PubMed2003

An intelligent system for diagnosis of the heart valve diseases with wavelet packet neural networks.

In this paper, an intelligent system is presented for interpretation of the Doppler signals of the heart valve diseases based on the pattern recognition. This paper especially deals with combination of the feature extraction and classification from measured Doppler signal waveforms at the heart valve using the Doppler Ultrasound. Because of this, a wavelet packet neural network model developed by us is used. The model consists of two layers: wavelet and multi-layer perceptron. The wavelet layer is used for adaptive feature extraction in the time-frequency domain and is composed of wavelet packet decomposition and wavelet packet entropy. The multi-layer perceptron used for classification is a feed-forward neural network. The performance of the developed system has been evaluated in 215 samples. The test results showed that this system was effective in detecting Doppler heart sounds. The correct classification rate was about 94% for abnormal and normal subjects.

Artificial Intelligence↗
PubMedDate not supplied

Building a relational database for a physician document index.

We show how three existing medical knowledge bases: Medical Subject Headings (MeSH), Standardized Nomenclature of Medicine (SNOMED) and Current Medical Information and Technology (CMIT) are mapped into a relational data model and stored on an Apollo workstation and an Intelligent Database Machine. Since two of these knowledge bases have been used in the indexing of medical literature and patient records, they can be useful not only as direct views on the organization of medical concepts but also as tools for the retrieval of documents. In order that the concepts from one knowledge base can be connected to those of the other knowledge base, a method has been developed for the semi-automatic merging of MeSH, SNOMED and CMIT. This method takes advantage of the relational model and the synonyms that are given in SNOMED and CMIT, in order to recommend concepts to be merged. An expert interacts with the system to accept or reject the recommendations of the computer. The method would apply equally well to other knowledge bases and is particularly well-suited for knowledge bases that contain tens of thousands of concepts.

Abstracting and Indexing↗
PubMed1996

Building intelligent alarm systems by combining mathematical models and inductive machine learning techniques.

In this article a technique is described to develop knowledge-based alarm systems for ventilator therapy, using mathematical modeling and machine learning. With a mathematical model airway pressure, expiratory gas flow and CO2 concentration at the endotracheal tube are simulated for patients, undergoing volume-controlled ventilation with constant ventilator settings, during normal functioning of the breathing circuit and during breathing circuit mishaps (leaks and obstructions). Simulations were performed for 94 physiologically different 'patients', by varying airway resistance and lung/thorax compliance values in the model. Each simulated breath was described by a set of derived signal features and a label that constituted during which event (normal function or mishap) the breath was recorded. With an inductive machine learning algorithm rules, linking signal feature values to breathing circuit events, were created from data of 54 of the simulated patients. The resulting set of rules was able to classify 99% of events in the data of the remaining 40 patients correctly. Of signals, measured at a ventilated lung simulator, 100% of events were classified correctly.

Airway Resistance↗
PubMed2004

Asymptotic properties of the Fisher kernel.

This letter analyzes the Fisher kernel from a statistical point of view. The Fisher kernel is a particularly interesting method for constructing a model of the posterior probability that makes intelligent use of unlabeled data (i.e., of the underlying data density). It is important to analyze and ultimately understand the statistical properties of the Fisher kernel. To this end, we first establish sufficient conditions that the constructed posterior model is realizable (i.e., it contains the true distribution). Realizability immediately leads to consistency results. Subsequently, we focus on an asymptotic analysis of the generalization error, which elucidates the learning curves of the Fisher kernel and how unlabeled data contribute to learning. We also point out that the squared or log loss is theoretically preferable-because both yield consistent estimators-to other losses such as the exponential loss, when a linear classifier is used together with the Fisher kernel. Therefore, this letter underlines that the Fisher kernel should be viewed not as a heuristics but as a powerful statistical tool with well-controlled statistical properties.

Algorithms↗
PubMed2004

Body area network--a key infrastructure element for patient-centered telemedicine.

The Body Area Network (BAN) extends the range of existing wireless network technologies by an ultra-low range, ultra-low power network solution optimised for long-term or continuous healthcare applications. It enables wireless radio communication between several miniaturised, intelligent Body Sensor (or actor) Units (BSU) and a single Body Central Unit (BCU) worn at the human body. A separate wireless transmission link from the BCU to a network access point--using different technology--provides for online access to BAN components via usual network infrastructure. The BAN network protocol maintains dynamic ad-hoc network configuration scenarios and co-existence of multiple networks.BAN is expected to become a basic infrastructure element for electronic health services: By integrating patient-attached sensors and mobile actor units, distributed information and data processing systems, the range of medical workflow can be extended to include applications like wireless multi-parameter patient monitoring and therapy support. Beyond clinical use and professional disease management environments, private personal health assistance scenarios (without financial reimbursement by health agencies / insurance companies) enable a wide range of applications and services in future pervasive computing and networking environments.

Artificial Intelligence↗
PubMed1996

An automated intelligent diagnostic system for the interpretation of umbilical artery Doppler velocimetry.

The objective is to develop an automated intelligent diagnostic system for the interpretation of umbilical artery velocity waveforms. An ultrasound instrument with pulsed-wave Doppler is connected to a microcomputer by means of a frame grabber. After data acquisition, umbilical Doppler velocimetry is handled as a pattern recognition (feature extraction and classification) and decision-making problem. Automated image processing (enhancement, smoothing/ thresholding and edge detection) and analysis are used for feature extraction. Six waveform indices obtained by feature extraction are used as input layer to vector quantization which classifies waveforms into six groups. A clinical decision is assigned to each group by the medical expert. Our system is trained by 278 and 380 waveform images of 94 normal and 157 high risk pregnancies, respectively. The system was tested with 193 and 61 images of normal and risky pregnancies; it was demonstrated that sensitivity and specificity of the system are 54.1% and 80.3%, respectively.

Artificial Intelligence↗
PubMed2004

An automated tissue preclassification approach for telepathology: implementation and performance analysis.

Telepathology is generally defined as the use of telecommunications technologies in the practice of anatomic or surgical pathology. In the usual telepathology scenario, a remotely located pathologist views images of tissues samples in order to render a diagnosis of the biopsy. Some telepathology systems involve interactive remote control of a microscope-based imaging system which delivers diagnostic quality imagery to the remote pathologist. The usefulness of such interactive systems depends on minimizing the end-to-end delays involved in controlling the robotic microscope, manipulating the tissue sample, and acquiring and transmitting the high-resolution image. An approach to minimizing end-to-end delay involves adding "intelligence" to the image acquisition system so that it can gather, classify, rank, and transmit diagnostically useful images in a semiautonomous fashion. In this research, we develop image analysis and ranking techniques which can improve the end-to-end performance of a robotic telepathology imaging system. Our semiautonomous image collection system uses morphological techniques to extract seed points for suspicious regions, a novel region growing algorithm to segment the regions of interest, and heuristically motivated expert system ranking techniques to select diagnostically relevant "next-step" image acquisitions. Diagnostic relevance of our segmentation and ranking algorithms is established via subjective and objective testing of the system. In subjective testing, pathologists Agree or Strongly Agree that all segmented regions are diagnostically relevant with probability greater than 0.75. In objective testing, 84% of "next-step" images acquired by our algorithms coincide with the areas most likely to be chosen by a pathologist.

Artificial Intelligence↗
PubMedDate not supplied

Concept for an intelligent anaesthesia EEG monitor.

Considering the fundamental difficulties to define the term 'depth of anaesthesia', a more feasible concept for assessment of 'adequacy of anaesthesia' will be explained. The basic requirements for a monitoring index are definite response, gradual scaling and independence from the anaesthetic technique used. Additionally the index should be predictive for appearance of clinical signs of an inadequate anaesthesia. Different signal-processing methods will be discussed to extract the relevant information from both the spontaneous and the evoked brain electrical activity. In this context well established methods like spectral analysis are investigated in combination with new and more sophisticated methods like bispectral analysis or wavelet decomposition. Since no single-parameter index has been defined for monitoring depth of anaesthesia, a set of EEG parameters may be more useful to take into account intra- and interindividual variability. In parallel to the description of the monitor concept, the investigation of neural nets and fuzzy techniques, in addition to or in substitution of conventional statistical methods, will be introduced. Examples are given for data quality assessment, parameter extraction and re-classification.

Algorithms↗
PubMed1997

Human factors in modern traffic systems.

Traffic systems are undergoing enormous change with the advent of Intelligent Transport Systems (ITS). Although productivity and quality of mobility are emerging interests, safety remains the predominant preoccupation of ITS human factors. It should be evident that while intelligent technologies may have the potential to improve traffic safety, they also have the potential to adversely affect it. Ultimately, the effect on safety depends on the specific technologies that are invoked and the manner in which they are incorporated within the vehicle as well as within the larger road transportation system. Current automotive developments can be characterized as technology-centred solutions rather than user-centred solutions. Greater effort must be directed at understanding and accommodating the human element in the road transportation system in order that future transportation objectives can be achieved. There is a need to expand the scope of traditional human factors to include macro-level effects as well as to place greater emphasis on understanding human interactions with other elements of the system. There is also increasing recognition of the urgent need for systematic procedures and criteria for testing the safety of ITS prior to large-scale market penetration.

Accidents, Traffic↗
PubMed2005

Temporal data mining for the quality assessment of hemodialysis services.

OBJECTIVE: This paper describes the temporal data mining aspects of a research project that deals with the definition of methods and tools for the assessment of the clinical performance of hemodialysis (HD) services, on the basis of the time series automatically collected during hemodialysis sessions. METHODS: Intelligent data analysis and temporal data mining techniques are applied to gain insight and to discover knowledge on the causes of unsatisfactory clinical results. In particular, two new methods for association rule discovery and temporal rule discovery are applied to the time series. Such methods exploit several pre-processing techniques, comprising data reduction, multi-scale filtering and temporal abstractions. RESULTS: We have analyzed the data of more than 5800 dialysis sessions coming from 43 different patients monitored for 19 months. The qualitative rules associating the outcome parameters and the measured variables were examined by the domain experts, which were able to distinguish between rules confirming available background knowledge and unexpected but plausible rules. CONCLUSION: The new methods proposed in the paper are suitable tools for knowledge discovery in clinical time series. Their use in the context of an auditing system for dialysis management helped clinicians to improve their understanding of the patients' behavior.

Algorithms↗
PubMed2004

Extending self-organizing particle systems to problem solving.

Self-organizing particle systems consist of numerous autonomous, purely reflexive agents ("particles") whose collective movements through space are determined primarily by local influences they exert upon one another. Inspired by biological phenomena (bird flocking, fish schooling, etc.), particle systems have been used not only for biological modeling, but also increasingly for applications requiring the simulation of collective movements such as computer-generated animation. In this research, we take some first steps in extending particle systems so that they not only move collectively, but also solve simple problems. This is done by giving the individual particles (agents) a rudimentary intelligence in the form of a very limited memory and a top-down, goal-directed control mechanism that, triggered by appropriate conditions, switches them between different behavioral states and thus different movement dynamics. Such enhanced particle systems are shown to be able to function effectively in performing simulated search-and-collect tasks. Further, computational experiments show that collectively moving agent teams are more effective than similar but independently moving ones in carrying out such tasks, and that agent teams of either type that split off members of the collective to protect previously acquired resources are most effective. This work shows that the reflexive agents of contemporary particle systems can readily be extended to support goal-directed problem solving while retaining their collective movement behaviors. These results may prove useful not only for future modeling of animal behavior, but also in computer animation, coordinated movement control in robotic teams, particle swarm optimization, and computer games.

Animals↗
PubMed2005

Computational intelligence-based optimisation of wastewater treatment plants.

Methods of computational intelligence (CI), especially fuzzy control and neuronal networks, are used for controlling and optimising of wastewater treatment plants. Areas of application are the control of sludge water dosage, of phosphate elimination by optimal precipitant dosage as well as an optimal aeration in the nitrification zone. In two municipal wastewater treatment plants with 60,000 and 12,600 person equivalents the controllers have been installed and optimised and they have been in operation for several years. Results of operation of the plants are presented in comparison to previously used classical control. Performance increased significantly and the outflow values could be kept securely below the government requirements without increase of the energy consumption. Peak loads in the inflow were eliminated in the plant and did not increase outflow concentrations. Results of operation for more than three years clearly show that the CI controller is a cost-efficient method for a sustainable rise of performance in municipal wastewater treatment plants.

Artificial Intelligence↗
PubMed2000

An evaluation of an intelligent home monitoring system.

A trial was performed of an intelligent monitoring system which used sensors in the home to identify emergencies by detecting deviations from normal activity patterns. The field trial lasted three months. Twenty-two elderly people agreed to participate. Their ages ranged from early 60s to over 85, with two-thirds in the age range 75-84 years. They lived in four different localities within the UK--Ipswich, Northumberland, Merseyside and Nottingham. A total of 61 alerts was recorded, at a mean frequency about one alert per month per client. Of the 61 alerts generated, 46 were classified as false alerts and the other 15 as genuine, although no real emergencies occurred during the study. Many people in the field trial reported enhanced feelings of safety and security, which could help to stimulate independence and help them to remain living in their own homes. The monitoring system increased the care choices available to elderly people and supported and enhanced the carer's role.

Aged↗
PubMed1996

The expert surgical assistant. An intelligent virtual environment with multimodal input.

Virtual Reality has made computer interfaces more intuitive but not more intelligent. This paper shows how an expert system can be coupled with multimodal input in a virtual environment to provide an intelligent simulation tool or surgical assistant. This is accomplished in three steps. First, voice and gestural input is interpreted and represented in a common semantic form. Second, a rule-based expert system is used to infer context and user actions from this semantic representation. Finally, the inferred user actions are matched against steps in a surgical procedure to monitor the user's progress and provide automatic feedback. In addition, the system can respond immediately to multimodal commands for navigational assistance and/or identification of critical anatomical structures. To show how these methods are used we present a prototype sinus surgery interface. The approach described here may easily be extended to a wide variety of medical and non-medical training applications by making simple changes to the expert system database and virtual environment models. Successful implementation of an expert system in both simulated and real surgery has enormous potential for the surgeon both in training and clinical practice.

Artificial Intelligence↗
PubMed2004

Evaluation of KNAVE-II: a tool for intelligent query and exploration of patient data.

We present the results of a preliminary evaluation of KNAVE-II, a distributed knowledge-based computational framework for visualization, interpretation, and exploration of longitudinal clinical data and of multiple levels of concepts derivable from these data. KNAVE-II uses a distributed architecture to access at run-time clinical time-oriented data, a domain-specific knowledge base containing properties of the clinical data, and a knowledge-based problem-solving method for computing on-the-fly interpretations of these data. The purpose of the evaluation was to compare efficiency and user satisfaction when answering clinical queries of variable complexity about clinical time-oriented data using KNAVE-II, versus using methods available in standard clinical settings: paper chart or electronic spreadsheet (ESS). Subjects answered high-complexity queries significantly faster using KNAVE-II than when using paper or ESS. User satisfaction with KNAVE-II was significantly superior compared to satisfaction using paper or ESS, based on a standard usability scale. Users also explicitly ranked KNAVE-II as superior to paper and the ESS.

Artificial Intelligence↗
PubMed2006

Computational intelligence for the detection and classification of malignant lesions in screening mammography.

This report deals with the discussion of the findings obtained from the application of two computational intelligence methodologies for the detection of microcalcifications in screening mammography data. Genetic programming and inductive machine learning have been applied, in order to produce meaningful diagnostic rules for the medical staff. The data used in the experiments correspond to information acquired from two images of each breast of the patient, along with some associated patient information such as the age at time of study. Similar datasets have been previously used in an attempt to facilitate the development of computer algorithms to aid screening. Experienced screening radiologists have double-read the screening mammograms, they have weighted the malignancy ratings and averaged out the levels of suspiciousness assigned to each finding in the screenings. The diagnostic rules which were obtained from both genetic programming and machine learning have been evaluated in detail and then analyzed and discussed by collaborative medical experts, in parallel to findings from related literature. Results seem encouraging for further use and analysis by medical staff specializing in screening mammography.

Artificial Intelligence↗
PubMed2006

Intelligent adaptive nonlinear flight control for a high performance aircraft with neural networks.

This paper describes the development of a neural network (NN) based adaptive flight control system for a high performance aircraft. The main contribution of this work is that the proposed control system is able to compensate the system uncertainties, adapt to the changes in flight conditions, and accommodate the system failures. The underlying study can be considered in two phases. The objective of the first phase is to model the dynamic behavior of a nonlinear F-16 model using NNs. Therefore a NN-based adaptive identification model is developed for three angular rates of the aircraft. An on-line training procedure is developed to adapt the changes in the system dynamics and improve the identification accuracy. In this procedure, a first-in first-out stack is used to store a certain history of the input-output data. The training is performed over the whole data in the stack at every stage. To speed up the convergence rate and enhance the accuracy for achieving the on-line learning, the Levenberg-Marquardt optimization method with a trust region approach is adapted to train the NNs. The objective of the second phase is to develop intelligent flight controllers. A NN-based adaptive PID control scheme that is composed of an emulator NN, an estimator NN, and a discrete time PID controller is developed. The emulator NN is used to calculate the system Jacobian required to train the estimator NN. The estimator NN, which is trained on-line by propagating the output error through the emulator, is used to adjust the PID gains. The NN-based adaptive PID control system is applied to control three angular rates of the nonlinear F-16 model. The body-axis pitch, roll, and yaw rates are fed back via the PID controllers to the elevator, aileron, and rudder actuators, respectively. The resulting control system has learning, adaptation, and fault-tolerant abilities. It avoids the storage and interpolation requirements for the too many controller parameters of a typical flight control system. Performance of the control system is successfully tested by performing several six-degrees-of-freedom nonlinear simulations.

Aircraft↗