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 1,063 records · Page 59Linked to original sources

PubMed2005

Web content management by self-organization.

We present a new method for content management and knowledge discovery using a topology-preserving neural network. The method, termed topological organization of content (TOC), can generate a taxonomy of topics from a set of unannotated, unstructured documents. The TOC consists of a hierarchy of self-organizing growing chains (GCs), each of which can develop independently in terms of size and topics. The dynamic development process is validated continuously using a proposed entropy-based Bayesian information criterion (BIC). Each chain meeting the criterion spans child chains, with reduced vocabularies and increased specializations. This results in a topological tree hierarchy, which can be browsed like a table of contents directory or web portal. A brief review is given on existing methods for document clustering and organization, and clustering validation measures. The proposed approach has been tested and compared with several existing methods on real world web page datasets. The results have clearly demonstrated the advantages and efficiency in content organization of the proposed method in terms of computational cost and representation. The TOC can be easily adapted for large-scale applications. The topology provides a unique, additional feature for retrieving related topics and confining the search space.

Abstracting and Indexing↗

PubMed2006

Ethanol sensitivity: a central role for CREB transcription regulation in the cerebellum.

BACKGROUND: Lowered sensitivity to the effects of ethanol increases the risk of developing alcoholism. Inbred mouse strains have been useful for the study of the genetic basis of various drug addiction-related phenotypes. Inbred Long-Sleep (ILS) and Inbred Short-Sleep (ISS) mice differentially express a number of genes thought to be implicated in sensitivity to the effects of ethanol. Concomitantly, there is evidence for a mediating role of cAMP/PKA/CREB signalling in aspects of alcoholism modelled in animals. In this report, the extent to which CREB signalling impacts the differential expression of genes in ILS and ISS mouse cerebella is examined. RESULTS: A training dataset for Machine Learning (ML) and Exploratory Data Analyses (EDA) was generated from promoter region sequences of a set of genes known to be targets of CREB transcription regulation and a set of genes whose transcription regulations are potentially CREB-independent. For each promoter sequence, a vector of size 132, with elements characterizing nucleotide composition features was generated. Genes whose expressions have been previously determined to be increased in ILS or ISS cerebella were identified, and their CREB regulation status predicted using the ML scheme C4.5. The C4.5 learning scheme was used because, of four ML schemes evaluated, it had the lowest predicted error rate. On an independent evaluation set of 21 genes of known CREB regulation status, C4.5 correctly classified 81% of instances with F-measures of 0.87 and 0.67 respectively for the CREB-regulated and CREB-independent classes. Additionally, six out of eight genes previously determined by two independent microarray platforms to be up-regulated in the ILS or ISS cerebellum were predicted by C4.5 to be transcriptionally regulated by CREB. Furthermore, 64% and 52% of a cross-section of other up-regulated cerebellar genes in ILS and ISS mice, respectively, were deemed to be CREB-regulated. CONCLUSION: These observations collectively suggest that ethanol sensitivity, as it relates to the cerebellum, may be associated with CREB transcription activity.

Alcoholism↗

PubMed1995

Representing clinical narratives using conceptual graphs.

The analysis of medical narratives and the generation of natural language expressions are strongly dependent on the existence of an adequate representation language. Such a language has to be expressive enough in order to handle the complexity of human reasoning in the domain. Sowa's Conceptual Graphs (CG) are an answer, and this paper presents a multilingual implementation, using French, English and German. Current developments demonstrate the feasibility of an approach to natural Language Understanding where semantic aspects are dominant, in contrast to syntax driven methods. The basic idea is to aggregate blocks of words according to semantic compatibility rules, following a method called Proximity Processing. The CG representation is gradually built, starting from single words in a semantic lexicon, to finally give a complete representation of the sentence under the form of a single CG. The process is dependent on specific rules of the medical domain, and for this reason is largely controlled by the declarative knowledge of the medical Linguistic Knowledge Base.

Artificial Intelligence↗

PubMed2007

Two-sample comparison based on prediction error, with applications to candidate gene association studies.

To take advantage of the increasingly available high-density SNP maps across the genome, various tests that compare multilocus genotypes or estimated haplotypes between cases and controls have been developed for candidate gene association studies. Here we view this two-sample testing problem from the perspective of supervised machine learning and propose a new association test. The approach adopts the flexible and easy-to-understand classification tree model as the learning machine, and uses the estimated prediction error of the resulting prediction rule as the test statistic. This procedure not only provides an association test but also generates a prediction rule that can be useful in understanding the mechanisms underlying complex disease. Under the set-up of a haplotype-based transmission/disequilibrium test (TDT) type of analysis, we find through simulation studies that the proposed procedure has the correct type I error rates and is robust to population stratification. The power of the proposed procedure is sensitive to the chosen prediction error estimator. Among commonly used prediction error estimators, the .632+ estimator results in a test that has the best overall performance. We also find that the test using the .632+ estimator is more powerful than the standard single-point TDT analysis, the Pearson's goodness-of-fit test based on estimated haplotype frequencies, and two haplotype-based global tests implemented in the genetic analysis package FBAT. To illustrate the application of the proposed method in population-based association studies, we use the procedure to study the association between non-Hodgkin lymphoma and the IL10 gene.

Adult↗

PubMed2005

Optimizing the automatic segmentation of the left ventricle in magnetic resonance images.

Automatic segmentation of the left ventricular (LV) myocardial borders in cardiovascular MR (CMR) images allows a significant speed-up of the procedure of quantifying LV function, and improves its reproducibility. The automated boundary delineation is usually based on a set of parameters that define the algorithms. Since the automatic segmentation algorithms are usually sensitive to the image quality and frequently depend heavily on the acquisition protocol, optimizing the parameters of the algorithm for such different protocols may be necessary to obtain optimal results. In other words, using a default set of parameters may be far from optimal for different scanners or protocols. For the MASS-software, for example, this means that a total of 14 parameters need to be optimized. This optimization is a difficult and labor-intensive process. To be able to more consistently and rapidly tune the parameters, an automated optimization system would be extremely desirable. In this paper we propose such an approach, which is based on genetic algorithms (GAs). The GA is an unsupervised iterative tool that generates new sets of parameters and converges toward an optimal set. We implemented and compared two different types of the genetic algorithms: a simple GA (SGA) and a steady state GA (2SGA). The difference between these two algorithms lies in the characteristics of the generated populations: "nonoverlapping populations" and "overlapping populations," respectively "nonoverlapping" population means that the two populations are disjoint, and "overlapping" means that the best parameters found in the previous generation are included in the present population. The performance of both algorithms was evaluated on twenty routinely obtained short-axis examinations (eleven examinations acquired with a steady-state free precession pulse sequence, and nine examinations with a gradient echo pulse sequence). The optimal parameters obtained with the GAs were used for the LV myocardial border delineation. Finally, the automatically outlined contours were compared to the gold standard--manually drawn contours by experts. The result of the comparison was expressed as a degree of similarity after a processing time of less than 72 h to a 59.5% of degree of similarity for SGA and a 66.7% of degree of similarity for 2SGA. In conclusion, genetic algorithms are very suitable to automatically tune the parameters of a border detection algorithm. Based on our data, the 2SGA was more suitable than the SGA method. This approach can be generalized to other optimization problems in medical image processing.

Algorithms↗

PubMed2002

Gene expression data analysis of human lymphoma using support vector machines and output coding ensembles.

The large amount of data generated by DNA microarrays was originally analysed using unsupervised methods, such as clustering or self-organizing maps. Recently supervised methods such as decision trees, dot-product support vector machines (SVM) and multi-layer perceptrons (MLP) have been applied in order to classify normal and tumoural tissues. We propose methods based on non-linear SVM with polynomial and Gaussian kernels, and output coding (OC) ensembles of learning machines to separate normal from malignant tissues, to classify different types of lymphoma and to analyse the role of sets of coordinately expressed genes in carcinogenic processes of lymphoid tissues. Using gene expression data from "Lymphochip", a specialised DNA microarray developed at Stanford University School of Medicine, we show that SVM can correctly separate normal from tumoural tissues, and OC ensembles can be successfully used to classify different types of lymphoma. Moreover, we identify a group of coordinately expressed genes related to the separation of two distinct subgroups inside diffuse large B-cell lymphoma (DLBCL), validating a previous Alizadeh's hypothesis about the existence of two distinct diseases inside DLBCL.

Artificial Intelligence↗

PubMedDate not supplied

Lines of curvature for polyp detection in virtual colonoscopy.

Computer-aided diagnosis (CAD) is a helpful addition to laborious visual inspection for preselection of suspected colonic polyps in virtual colonoscopy. Most of the previous work on automatic polyp detection makes use of indicators based on the scalar curvature of the colon wall and can result in many false-positive detections. Our work tries to reduce the number of false-positive detections in the preselection of polyp candidates. Polyp surface shape can be characterized and visualized using lines of curvature. In this paper, we describe techniques for generating and rendering lines of curvature on surfaces and we show that these lines can be used as part of a polyp detection approach. We have adapted existing approaches on explicit triangular surface meshes, and developed a new algorithm on implicit surfaces embedded in 3D volume data. The visualization of shaded colonic surfaces can be enhanced by rendering the derived lines of curvature on these surfaces. Features strongly correlated with true-positive detections were calculated on lines of curvature and used for the polyp candidate selection. We studied the performance of these features on 5 data sets that included 331 pre-detected candidates, of which 50 sites were true polyps. The winding angle had a significant discriminating power for true-positive detections, which was demonstrated by a Wilcoxon rank sum test with p < 0.001. The median winding angle and inter-quartile range (IQR) for true polyps were 7.817 and 6.770 - 9.288 compared to 2.954 and 1.995 - 3.749 for false-positive detections.

Algorithms↗

PubMed2006

Dynamic reasoning to solve complex problems in activated sludge processes: a step further in decision support systems.

Decision support systems (DSS) have generated high expectations as a tool to support activated sludge operation because of their ability to represent heuristic reasoning and to handle large amounts of qualitative, uncertain and low-accuracy data. Previous applications have been satisfactory to control simple problems, when static reasoning and literature-based solutions were enough. However to face complex operational problems with biological origin and slow dynamics (e.g. solids separation problems), it is necessary to use dynamic reasoning and apply long-term control strategies, monitoring the evolution of the process and adjusting the action plan according to the feed back of the process. This paper presents a dynamic reasoning DSS to face solids separation problems in the activated sludge system. The DSS is capable of identifying the complex problem affecting the process, determining if the current situation is new or a continuation from the previous one, assessing what is the specific cause of the situation, and recommending a long-term control strategy, which is daily adjusted according to the evolution of the process.

Artificial Intelligence↗

PubMed2006

The fundamentals of average local variance--Part I: Detecting regular patterns.

The method of average local variance (ALV) computes the mean of the standard deviation values derived for a 3 x 3 moving window on a successively coarsened image to produce a function of ALV versus spatial resolution. In developing ALV, the authors used approximately a doubling of the pixel size at each coarsening of the image. They hypothesized that ALV is low when the pixel size is smaller than the size of scene objects because the pixels on the object will have similar response values. When the pixel and objects are of similar size, they will tend to vary in response and the ALV values will increase. As the size of pixels increase further, more objects will be contained in a single pixel and ALV will decrease. The authors showed that various cover types produced single peak ALV functions that inexplicitly peaked when the pixel size was 1/2 to 3/4 of the object size. This paper reports on work done to explore the characteristics of the various forms of the ALV function and to understand the location of the peaks that occur in this function. The work was conducted using synthetically generated image data. The investigation showed that the hypothesis originally proposed in is not adequate. A new hypothesis is proposed that the ALV function has peak locations that are related to the geometric size of pattern structures in the scene. These structures are not always the same as scene objects. Only in cases where the size of and separation between scene objects are equal does the ALV function detect the size of the objects. In situations where the distance between scene objects are larger than their size, the ALV function has a peak at the object separation, not at the object size. This work has also shown that multiple object structures of different sizes and distances in the image provide multiple peaks in the ALV function and that some of these structures are not implicitly recognized as such from our perspective. However, the magnitude of these peaks depends on the response mix in the structures, complicating their interpretation and analysis. The analysis of the ALV Function is, thus, more complex than that generally reported in the literature.

Algorithms↗

PubMed2005

Recognition and verification of unconstrained handwritten words.

This paper presents a novel approach for the verification of the word hypotheses generated by a large vocabulary, offline handwritten word recognition system. Given a word image, the recognition system produces a ranked list of the N-best recognition hypotheses consisting of text transcripts, segmentation boundaries of the word hypotheses into characters, and recognition scores. The verification consists of an estimation of the probability of each segment representing a known class of character. Then, character probabilities are combined to produce word confidence scores which are further integrated with the recognition scores produced by the recognition system. The N-best recognition hypothesis list is reranked based on such composite scores. In the end, rejection rules are invoked to either accept the best recognition hypothesis of such a list or to reject the input word image. The use of the verification approach has improved the word recognition rate as well as the reliability of the recognition system, while not causing significant delays in the recognition process. Our approach is described in detail and the experimental results on a large database of unconstrained handwritten words extracted from postal envelopes are presented.

Algorithms↗

PubMed2005

Monitoring and fault diagnosis of hybrid systems.

Many networked embedded sensing and control systems can be modeled as hybrid systems with interacting continuous and discrete dynamics. These systems present significant challenges for monitoring and diagnosis. Many existing model-based approaches focus on diagnostic reasoning assuming appropriate fault signatures have been generated. However, an important missing piece is the integration of model-based techniques with the acquisition and processing of sensor signals and the modeling of faults to support diagnostic reasoning. This paper addresses key modeling and computational problems at the interface between model-based diagnosis techniques and signature analysis to enable the efficient detection and isolation of incipient and abrupt faults in hybrid systems. A hybrid automata model that parameterizes abrupt and incipient faults is introduced. Based on this model, an approach for diagnoser design is presented. The paper also develops a novel mode estimation algorithm that uses model-based prediction to focus distributed processing signal algorithms. Finally, the paper describes a diagnostic system architecture that integrates the modeling, prediction, and diagnosis components. The implemented architecture is applied to fault diagnosis of a complex electro-mechanical machine, the Xerox DC265 printer, and the experimental results presented validate the approach. A number of design trade-offs that were made to support implementation of the algorithms for online applications are also described.

Algorithms↗

PubMedDate not supplied

Intelligent medical information filtering.

This paper describes an intelligent information filtering system to assist users to be notified of updates to new and relevant medical information. Among the major problems users face is the large volume of medical information that is generated each day, and the need to filter and retrieve relevant information. The Internet has dramatically increased the amount of electronically accessible medical information and reduced the cost and time needed to publish. The opportunity of the Internet for the medical profession and consumers is to have more information to make decisions and this could potentially lead to better medical decisions and outcomes. However, without the assistance from professional medical librarians, retrieving new and relevant information from databases and the Internet remains a challenge. Many physicians do not have access to the services of a medical librarian. Most physicians indicate on surveys that they do not prefer to retrieve the literature themselves, or visit libraries because of the lack of recent materials, poor organisation and indexing of materials, lack of appropriate and available material, and lack of time. The information filtering system described in this paper records the online web browsing behaviour of each user and creates a user profile of the index terms found on the web pages visited by the user. A relevance-ranking algorithm then matches the user profiles to the index terms of new health care web pages that are added each day. The system creates customised summaries of new information for each user. A user can then connect to the web site to read the new information. Relevance feedback buttons on each page ask the user to rate the usefulness of the page to their immediate information needs. Errors in relevance ranking are reduced in this system by having both the user profile and medical information represented in the same representation language using a controlled vocabulary. This system also updates the user profiles, automatically relieving this burden from the user, but also allowing the user to explicitly state preferences. An initial evaluation of this system was done with health consumers using a web site on consumer health. It was found that users often modified their criteria for what they considered relevant not only between browsing sessions but also during a session. A user's criteria for what is relevant is constantly changing as they interact with the information. New revised metrics of recall and precision are needed to account for the partially relevant judgements and the dynamically changing criteria of users. Future research, development, and evaluation of interactive information retrieval systems will need to take into account the users' dynamically changing criteria of relevance.

Artificial Intelligence↗

PubMed2005

Detection of malignancy in cytology specimens using spectral-spatial analysis.

Despite low sensitivity (around 60%), cytomorphologic examination of urine specimens represents the standard procedure in the diagnosis and follow-up of bladder cancer. Although color is information-rich, morphologic diagnoses are rendered almost exclusively on the basis of spatial information. We hypothesized that quantitative assessment of color (more precisely, of spectral properties) using liquid crystal-based spectral fractionation, combined with genetic algorithm-based spatial analysis, can improve the accuracy of traditional cytologic examination. Images of various cytological specimens were collected every 10 nm from 400 to 700 nm to create an image stack. The resulting data sets were analyzed using the Los Alamos-developed GENetic Imagery Exploitation (GENIE) package, a hybrid genetic algorithm that segments (classifies) images using automatically 'learned' spatio-spectral features. In an evolutionary fashion, GENIE generates a series of algorithms or 'chromosomes', keeping the one with best fitness with respect to a user-defined training set. First, we tested the system to determine if it could recognize malignant cells using artificial cytology specimens constructed to completely avoid the requirement for human interpretation. GENIE was able to differentiate malignant from benign cells and to estimate their relative proportions in controlled mixtures. We then tested the system on routine cytology specimens. When targeted to detect malignant urothelial cells in cytology specimens, GENIE showed a combined sensitivity and specificity of 85 and 95%, in samples drawn from two separate institutions over a span of 4 years. When trained on cases initially diagnosed as 'atypical' but with unequivocal follow-up by biopsy, surgical specimen or cytology, GENIE showed efficiency superior to the cytopathologist with respect to predicting the follow-up result in a cohort of 85 cases. We believe that, in future, this type of methodology could be used as an ancillary test in cytopathology, in a manner analogous to immunostaining, in those situations when a definitive diagnosis cannot be rendered based solely on the morphology.

Adenocarcinoma↗

PubMed2006

Bayesian detection of periodic mRNA time profiles without use of training examples.

BACKGROUND: Detection of periodically expressed genes from microarray data without use of known periodic and non-periodic training examples is an important problem, e.g. for identifying genes regulated by the cell-cycle in poorly characterised organisms. Commonly the investigator is only interested in genes expressed at a particular frequency that characterizes the process under study but this frequency is seldom exactly known. Previously proposed detector designs require access to labelled training examples and do not allow systematic incorporation of diffuse prior knowledge available about the period time. RESULTS: A learning-free Bayesian detector that does not rely on labelled training examples and allows incorporation of prior knowledge about the period time is introduced. It is shown to outperform two recently proposed alternative learning-free detectors on simulated data generated with models that are different from the one used for detector design. Results from applying the detector to mRNA expression time profiles from S. cerevisiae showsthat the genes detected as periodically expressed only contain a small fraction of the cell-cycle genes inferred from mutant phenotype. For example, when the probability of false alarm was equal to 7%, only 12% of the cell-cycle genes were detected. The genes detected as periodically expressed were found to have a statistically significant overrepresentation of known cell-cycle regulated sequence motifs. One known sequence motif and 18 putative motifs, previously not associated with periodic expression, were also over represented. CONCLUSION: In comparison with recently proposed alternative learning-free detectors for periodic gene expression, Bayesian inference allows systematic incorporation of diffuse a priori knowledge about, e.g. the period time. This results in relative performance improvements due to increased robustness against errors in the underlying assumptions. Results from applying the detector to mRNA expression time profiles from S. cerevisiae include several new findings that deserve further experimental studies.

Algorithms↗

PubMed2002

Knowledge-based assessment of gene expression data from chemiluminescence detection.

The first problem in gene expression profiling to be solved is choosing the appropriate gene array, detection procedure, image analysis and data generation depending on the organism of interest, equipment and budget. The next one is how to deduce biologically meaningful data. We assessed gene expression data from chemiluminescent detection and empirically found criteria for the reliable identification of biologically meaningful expression ratios. Current statistical assessments are often applied unreflectedly concerning problems occurring in practice. So interesting results are considered to be irrelevant. This requires a laborious data check. We suggest automation. Our empirically found criteria were transformed into and validated by a knowledge-based system. This system is adaptable to all other methods of expression profiling. We compared the experience-based and new knowledge-based assessment of the expression data from our chemiluminescent and additionally radioactive detection of several experiments with published data to evaluate our entire procedure. With our adaptation of chemiluminescence detection to commercially available Escherichia coli gene arrays we present a useful alternative to common procedures in gene expression monitoring. Moreover, with our consideration of plasmid-harbouring E. coli strains we provide the opportunity to monitor gene expression during processes requiring any plasmids (e.g. recombinant protein expression).

Artificial Intelligence↗

PubMed2005

Predictive neural networks for gene expression data analysis.

Gene expression data generated by DNA microarray experiments have provided a vast resource for medical diagnosis and disease understanding. Most prior work in analyzing gene expression data, however, focuses on predictive performance but not so much on deriving human understandable knowledge. This paper presents a systematic approach for learning and extracting rule-based knowledge from gene expression data. A class of predictive self-organizing networks known as Adaptive Resonance Associative Map (ARAM) is used for modelling gene expression data, whose learned knowledge can be transformed into a set of symbolic IF-THEN rules for interpretation. For dimensionality reduction, we illustrate how the system can work with a variety of feature selection methods. Benchmark experiments conducted on two gene expression data sets from acute leukemia and colon tumor patients show that the proposed system consistently produces predictive performance comparable, if not superior, to all previously published results. More importantly, very simple rules can be discovered that have extremely high diagnostic power. The proposed methodology, consisting of dimensionality reduction, predictive modelling, and rule extraction, provides a promising approach to gene expression analysis and disease understanding.

Animals↗

PubMed2001

Acceptance of rules generated by machine learning among medical experts.

OBJECTIVES: The aim was to evaluate the potential for monotonicity constraints to bias machine learning systems to learn rules that were both accurate and meaningful. METHODS: Two data sets, taken from problems as diverse as screening for dementia and assessing the risk of mental retardation, were collected and a rule learning system, with and without monotonicity constraints, was run on each. The rules were shown to experts, who were asked how willing they would be to use such rules in practice. The accuracy of the rules was also evaluated. RESULTS: Rules learned with monotonicity constraints were at least as accurate as rules learned without such constraints. Experts were, on average, more willing to use the rules learned with the monotonicity constraints. CONCLUSIONS: The analysis of medical databases has the potential of improving patient outcomes and/or lowering the cost of health care delivery. Various techniques, from statistics, pattern recognition, machine learning, and neural networks, have been proposed to "mine" this data by uncovering patterns that may be used to guide decision making. This study suggests cognitive factors make learned models coherent and, therefore, credible to experts. One factor that influences the acceptance of learned models is consistency with existing medical knowledge.

Alzheimer Disease↗

PubMed2004

Comparing different classifiers for automatic age estimation.

We describe a quantitative evaluation of the performance of different classifiers in the task of automatic age estimation. In this context, we generate a statistical model of facial appearance, which is subsequently used as the basis for obtaining a compact parametric description of face images. The aim of our work is to design classifiers that accept the model-based representation of unseen images and produce an estimate of the age of the person in the corresponding face image. For this application, we have tested different classifiers: a classifier based on the use of quadratic functions for modeling the relationship between face model parameters and age, a shortest distance classifier, and artificial neural network based classifiers. We also describe variations to the basic method where we use age-specific and/or appearance specific age estimation methods. In this context, we use age estimation classifiers for each age group and/or classifiers for different clusters of subjects within our training set. In those cases, part of the classification procedure is devoted to choosing the most appropriate classifier for the subject/age range in question, so that more accurate age estimates can be obtained. We also present comparative results concerning the performance of humans and computers in the task of age estimation. Our results indicate that machines can estimate the age of a person almost as reliably as humans.

Age Factors↗