SEARCH · PubMed Health

Results for “generative artificial intelligence”

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 451 records · Page 25Linked to original sources

PubMed2004

Thirty years of computational autopoiesis: a review.

Computational autopoiesis--the realization of autopoietic entities in computational media--holds an important and distinctive role within the field of artificial life. Its earliest formulation by Francisco Varela, Humberto Maturana, and Ricardo Uribe was seminal in demonstrating the use of an artificial, computational medium to explore the most basic question of the abstract nature of living systems--over a decade in advance of the first Santa Fe Workshop on Artificial Life. The research program it originated has generated substantive demonstrations of progressively richer, lifelike phenomena. It has also sharply illuminated both conceptual and methodological problems in the field. This article provides an integrative overview of the sometimes disparate work in this area, and argues that computational autopoiesis continues to provide an effective framework for addressing key open problems in artificial life.

Artificial Intelligence↗

PubMed2006

Adaptive force regulation of muscle strengthening rehabilitation device with magnetorheological fluids.

In rehabilitation from neuromuscular trauma or injury, strengthening exercises are often prescribed by physical therapists to recover as much function as possible. Strengthening equipment used in clinical settings range from low-cost devices, such as sandbag weights or elastic bands to large and expensive isotonic and isokinetic devices. The low-cost devices are incapable of measuring strength gains and apply resistance based on the lowest level of torque that is produced by a muscle group. Resistance that varies with joint angle can be achieved with isokinetic devices in which angular velocity is held constant and variable torque is generated when the patient attempts to move faster than the device but are ineffective if a patient cannot generate torque rapidly. In this paper, we report the development of a versatile rehabilitation device that can be used to strengthen different muscle groups based on the torque generating capability of the muscle that changes with joint angle. The device is low cost, is smaller than other commercially available machines, and can be programmed to apply resistance that is unique to a particular patient and that will optimize strengthening. The core of the device, a damper with smart magnetorheological fluids, provides passive exercise force. A digital adaptive control is capable of regulating exercise force precisely following the muscle strengthening profile prescribed by a physical therapist. The device could be programmed with artificial intelligence to dynamically adjust the target force profile to optimize rehabilitation effects. The device provides both isometric and isokinetic strength training and can be developed into a small, low-cost device that may be capable of providing optimal strengthening in the home.

Adaptation, Physiological↗

PubMed2006

Algorithms for automatic interpretation of high resolution mass spectra.

Automated interpretation of high-resolution mass spectra in a reliable and efficient manner represents a highly challenging computational problem. This work aims at developing methods for reducing a high-resolution mass spectrum into its monoisotopic peak list, and automatically assigning observed masses to known fragment ion masses if the protein sequence is available. The methods are compiled into a suite of data reduction algorithms which is called MasSPIKE (Mass Spectrum Interpretation and Kernel Extraction). MasSPIKE includes modules for modeling noise across the spectrum, isotopic cluster identification, charge state determination, separation of overlapping isotopic distributions, picking isotopic peaks, aligning experimental and theoretical isotopic distributions for estimating a monoisotopic peak's location, generating the monoisotopic mass list, and assigning the observed monoisotopic masses to possible protein fragments. The method is tested against a complex top-down spectrum of bovine carbonic anhydrase. Results of each of the individual modules are compared with previously published work.

Algorithms↗

PubMed2002

Of truth and pathways: chasing bits of information through myriads of articles.

Knowledge on interactions between molecules in living cells is indispensable for theoretical analysis and practical applications in modern genomics and molecular biology. Building such networks relies on the assumption that the correct molecular interactions are known or can be identified by reading a few research articles. However, this assumption does not necessarily hold, as truth is rather an emerging property based on many potentially conflicting facts. This paper explores the processes of knowledge generation and publishing in the molecular biology literature using modelling and analysis of real molecular interaction data. The data analysed in this article were automatically extracted from 50000 research articles in molecular biology using a computer system called GeneWays containing a natural language processing module. The paper indicates that truthfulness of statements is associated in the minds of scientists with the relative importance (connectedness) of substances under study, revealing a potential selection bias in the reporting of research results. Aiming at understanding the statistical properties of the life cycle of biological facts reported in research articles, we formulate a stochastic model describing generation and propagation of knowledge about molecular interactions through scientific publications. We hope that in the future such a model can be useful for automatically producing consensus views of molecular interaction data.

Algorithms↗

PubMed2005

Interpreting spatial and temporal neural activity through a recurrent neural network brain-machine interface.

We propose the use of optimized brain-machine interface (BMI) models for interpreting the spatial and temporal neural activity generated in motor tasks. In this study, a nonlinear dynamical neural network is trained to predict the hand position of primates from neural recordings in a reaching task paradigm. We first develop a method to reveal the role attributed by the model to the sampled motor, premotor, and parietal cortices in generating hand movements. Next, using the trained model weights, we derive a temporal sensitivity measure to asses how the model utilized the sampled cortices and neurons in real-time during BMI testing.

Algorithms↗

PubMed2001

Real-time anatomical 3D image extraction for laparoscopic surgery.

Progress in the application of augmented reality to laparoscopic surgery has been limited by the difficulty associated with generating geometric information about the current patient in real time. Structured light techniques are well known methods for generating range images using a camera and projector, but typically fail when faced with biological specimens. We describe techniques and equipment that have shown promise for acquisition of range images for use in a real-time augmented reality system for laparoscopic surgery.

Artificial Intelligence↗

PubMed1995

Enhancements of anatomical information in UMLS knowledge sources.

Although anatomical terminology forms a part of biomedical structured vocabularies, available sources lack the requisite granularity, semantic types and relationships for comprehensively and consistently representing anatomical concepts in machine readable form. Thoracic angiology was selected as a proof of concept experiment for in depth representation of symbolic information in gross anatomy through the enhancement of semantic types, concepts and relationships in UMLS. Provided the representation of concepts is comprehensive, hierarchies generated with four types of simple relationships are capable of displaying anatomical information from the systemic view point with sufficient detail to meet the needs of applications in basic science education and in the practice of surgical subspecialties.

Anatomy↗

PubMed1996

Design of a clinical event monitor.

The issues and implementation of a clinical event monitor are described. An event monitor generates messages for providers, patients, and organizations based on clinical events and patient data. For example, an order for a medication might trigger the generation of a warning about a drug interaction. A model based on the active database literature has as its main components an event (which triggers a rule to fire), a condition (which tests whether an action ought to be performed), and an action (often the generation of a message). The details of implementing such a monitor are described, using as an example the Columbia-Presbyterian Medical Center clinical event monitor, which is based on the Arden Syntax for Medical Logic Modules.

Artificial Intelligence↗

PubMed1997

Clinical-HINTS: integrated intelligent ICU patient monitoring and information management system.

Clinical-HINTS (Health Intelligence System) is a horizontally integrated decision support system (DSS) designed to meet the requirements for intelligent real-time clinical information management in critical care medical environments and to lay the foundation for the development of the next generation of intelligent medical instrumentation. The system presented was developed to refine and complement the information yielded by clinical laboratory investigations, thereby benefiting the management of the intensive care unit (ICU) patient. More specifically, Clinical-HINTS was developed to provide computer-based assistance with the acquisition, organisation and display, storage and retrieval, communication and generation of real-time patient-specific clinical information in an ICU. Clinical-HINTS is an object-oriented system developed in C+2 to run under Microsoft Windows as an embryo intelligent agent. Current generic reasoning skills include perception and reactive cognition of patient status but exclude therapeutic action. The system monitors the patient by communicating with the available sources of data and uses generic reasoning skills to generate intelligent alarms, or HINTS, on various levels of interpretation of an observed dysfunction, even in the presence of complex disorders. The system's communication and information management capabilities are used to acquire physiological data, and to store them along with their interpretations and any interventions for the dynamic recognition of interrelated pathophysiological states or clinical events.

Artificial Intelligence↗

PubMed2005

Motif extraction and protein classification.

We present a novel unsupervised method for extracting meaningful motifs from biological sequence data. This de novo motif extraction (MEX) algorithm is data driven, finding motifs that are not necessarily over-represented in the data. Applying MEX to the oxidoreductases class of enzymes, containing approximately 7000 enzyme sequences, a relatively small set of motifs is obtained. This set spans a motif-space that is used for functional classification of the enzymes by an SVM classifier. The classification based on MEX motifs surpasses that of two other SVM based methods: SVMProt, a method based on the analysis of physical-chemical properties of a protein generated from its sequence of amino acids, and SVM applied to a Smith-Waterman distances matrix. Our findings demonstrate that the MEX algorithm extracts relevant motifs, supporting a successful sequence-to-function classification.

Algorithms↗

PubMed2004

Synchrony detection and amplification by silicon neurons with STDP synapses.

Spike-timing dependent synaptic plasticity (STDP) is a form of plasticity driven by precise spike-timing differences between presynaptic and postsynaptic spikes. Thus, the learning rules underlying STDP are suitable for learning neuronal temporal phenomena such as spike-timing synchrony. It is well known that weight-independent STDP creates unstable learning processes resulting in balanced bimodal weight distributions. In this paper, we present a neuromorphic analog very large scale integration (VLSI) circuit that contains a feedforward network of silicon neurons with STDP synapses. The learning rule implemented can be tuned to have a moderate level of weight dependence. This helps stabilise the learning process and still generates binary weight distributions. From on-chip learning experiments we show that the chip can detect and amplify hierarchical spike-timing synchrony structures embedded in noisy spike trains. The weight distributions of the network emerging from learning are bimodal.

Action Potentials↗

PubMed2001

Simulating the growth of viruses.

To explore how the genome of an organism defines its growth, we have developed a computer simulation for the intracellular growth of phage T7 on its E. coli host. Our simulation, which incorporates 30 years of genetic, biochemical, physiological, and biophysical data, is used here to study how the intracellular resources of the host, determined by the specific growth rate of the host, contribute toward phage development. It is also used to probe how changes in the linear organization of genetic elements on the T7 genome can affect T7 development. Further, we show how time-series trajectories of T7 mRNA and protein levels generated by the simulation may be used as raw data to test data-mining strategies, specifically, to identify partners in protein-protein interactions. Finally, we suggest how generalization of this work can lead to a knowledge-driven simulation for the growth of any virus.

Algorithms↗

PubMed1995

Clinical simulation using context-sensitive temporal probability models.

We present a language for representing context-sensitive temporal probabilistic knowledge. Context constraints allow inference to be focused on only the relevant portions of the probabilistic knowledge. We provide a declarative semantics for our language and an implemented algorithm (BNG) that generates Bayesian networks to compute the posterior probabilities of queries. We illustrate the use of the BNG system by applying it to the problem of modeling the effects of medications and other interventions on the condition of a patient in cardiac arrest.

Artificial Intelligence↗

PubMed1990

Automated conformational analysis and structure generation: algorithms for molecular perception.

Many methodologies for performing automated conformational analysis require some means of "perceiving" a molecule to determine features of interest. Algorithms for finding rings, bond orders, and stereocenters and detecting the presence of substructural fragments have been developed. These algorithms are described, emphasizing their importance in conformational analysis.

Artificial Intelligence↗

PubMed2004

Development of a computational approach to predict blood-brain barrier permeability.

The objectives of this study were to generate a data set of blood-brain barrier (BBB) permeability values for drug-like compounds and to develop a computational model to predict BBB permeability from structure. The BBB permeability, expressed as permeability-surface area product (PS, quantified as logPS), was determined for 28 structurally diverse drug-like compounds using the in situ rat brain perfusion technique. A linear model containing three descriptors, logD, van der Waals surface area of basic atoms, and polar surface area, was developed based on 23 compounds in our data set, where the penetration across the BBB was assumed to occur primarily by passive diffusion. The correlation coefficient (R(2)) and standard deviation (S.D.) of the model-predicted logPS against the observed are 0.74 and 0.50, respectively. If an outlier was removed from the training data set, the R(2) and S.D. were 0.80 and 0.44, respectively. This new model was tested in two literature data sets, resulting in an R(2) of 0.77 to 0.94 and a S.D. of 0.38 to 0.51. For comparison, four literature models, logP, logD, log(D. MW(-0.5)), and linear free energy relationship, were tested using the set of 23 compounds primarily crossing the BBB by passive diffusion, resulting in an R(2) of 0.33 to 0.61 and a S.D. of 0.59 to 0.76. In summary, we have generated the largest PS data set and developed a robust three-descriptor model that can quantitatively predict BBB permeability. This model may be used in a drug discovery setting to predict the BBB permeability of new chemical entities.

Algorithms↗

PubMed2006

Cluster analysis of networks generated through homology: automatic identification of important protein communities involved in cancer metastasis.

BACKGROUND: Protein-protein interactions have traditionally been studied on a small scale, using classical biochemical methods to investigate the proteins of interest. More recently large-scale methods, such as two-hybrid screens, have been utilised to survey extensive portions of genomes. Current high-throughput approaches have a relatively high rate of errors, whereas in-depth biochemical studies are too expensive and time-consuming to be practical for extensive studies. As a result, there are gaps in our knowledge of many key biological networks, for which computational approaches are particularly suitable. RESULTS: We constructed networks, or 'interactomes', of putative protein-protein interactions in the rat proteome--the rat being an organism extensively used for cancer studies. This was achieved by integrating experimental protein-protein interaction data from many species and translating this data into the reference frame of the rat. The putative rat protein interactions were given confidence scores based on their homology to proteins that have been experimentally observed to interact. The confidence score was furthermore weighted according to the extent of the experimental evidence, giving a higher weight to more frequently observed interactions. The scoring function was subsequently validated and networks constructed around key proteins, identified as being highly up- or down-regulated in rat cell lines of high metastatic potential. Using clustering methods on the networks, we have identified key protein communities involved in cancer metastasis. CONCLUSION: The protein network generation and subsequent network analysis used here, were shown to be useful for highlighting key proteins involved in metastasis. This approach, in conjunction with microarray expression data, can be extended to other species, thereby suggesting possible pathways around proteins of interest.

Algorithms↗

PubMed2005

Extracting drug-drug interaction articles from MEDLINE to improve the content of drug databases.

Drug-drug interaction systems exhibit low signal-to-noise ratios because of the amount of clinically insignificant or inaccurate information they contain. MEDLINE represents a respected source of peer-reviewed biomedical citations that potentially might serve as a valuable source of drug-drug interaction information, if relevant articles could be pinpointed effectively and efficiently. We evaluated the classification capability of Support Vector Machines as a method for locating articles about drug interactions. We used a corpus of "positive" and"negative" drug interaction citations to generate datasets composed of MeSH terms, CUI-tagged title and abstract text, and stemmed text words. The study showed that automated classification techniques have the potential to perform at least as well as PubMed in identifying drug-drug interaction articles.

Algorithms↗

PubMed1997

The digital anatomist information system and its use in the generation and delivery of Web-based anatomy atlases.

Advances in network and imaging technology, coupled with the availability of 3-D datasets such as the Visible Human, provide a unique opportunity for developing information systems in anatomy that can deliver relevant knowledge directly to the clinician, researcher or educator. A software framework is described for developing such a system within a distributed architecture that includes spatial and symbolic anatomy information resources, Web and custom servers, and authoring and end-user client programs. The authoring tools have been used to create 3-D atlases of the brain, knee and thorax that are used both locally and throughout the world. For the one and a half year period from June 1995-January 1997, the on-line atlases were accessed by over 33,000 sites from 94 countries, with an average of over 4000 "hits" per day, and 25,000 hits per day during peak exam periods. The atlases have been linked to by over 500 sites, and have received at least six unsolicited awards by outside rating institutions. The flexibility of the software framework has allowed the information system to evolve with advances in technology and representation methods. Possible new features include knowledge-based image retrieval and tutoring, dynamic generation of 3-D scenes, and eventually, real-time virtual reality navigation through the body. Such features, when coupled with other on-line biomedical information resources, should lead to interesting new ways for managing and accessing structural information in medicine.

Anatomy↗