SEARCH · PubMed Health

Results for “Predictive Learning Models”

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 1,765 records · Page 98Linked to original sources

PubMed2026

Integrated transcriptome analysis and machine learning to construct a homeostatic model of acetylation for bladder cancer and validate the key gene CES1.

BACKGROUND: Bladder cancer (BLCA) is one of the most common malignant tumors of the urinary system. Protein acetylation (PA) plays a critical role in regulating multiple biological processes (BPs), cellular homeostasis, and cancer-related signaling pathways. This study aimed to construct a homeostatic model of acetylation for BLCA using integrated transcriptome analysis and machine learning and to validate the key gene CES1. METHODS: RNA sequencing (RNA-seq) and clinical data were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Acetylation-related differentially expressed genes (DEGs) in BLCA were screened using differential expression analysis (DEA). An acetylation homeostatic model was constructed via univariate, machine learning-based least absolute shrinkage and selection operator (LASSO) and multivariate Cox regression analyses, followed by validation in multiple cohorts. Single-cell RNA-seq analysis was used to explore gene expression patterns in diverse cell types. Enrichment analysis (EA), immune infiltration, and drug sensitivity analysis (DSA) were performed to characterize molecular features of different risk groups. Finally, the biological function of CES1 as the key gene was verified by in vitro knockdown experiments. RESULTS: We established a robust acetylation homeostatic model consisting of five genes, which effectively predicted overall survival (OS) and served as an independent prognostic factor in BLCA. High-risk patients showed significantly poorer prognosis, distinct immune infiltration profiles, and differential drug sensitivity. CES1 was identified and validated as the key gene in this model, which was highly expressed in BLCA and associated with poor prognosis. Knockdown of CES1 markedly suppressed cell proliferation, invasion, and migration, and reduced intracellular coenzyme A (CoA) levels, thereby regulating PA homeostasis. CONCLUSIONS: We developed and validated a novel acetylation homeostatic model for survival stratification and personalized treatment guidance in BLCA, based on integrated transcriptome analysis and machine learning. CES1 is closely associated with intracellular CoA levels and the malignant progression of BLCA. Its potential association with PA homeostasis requires further mechanistic validation, and it may act as a candidate therapeutic biomarker for BLCA.

Bladder cancer (BLCA)↗

PubMed1994

Nominal and physical decision criteria in same-different judgments.

We propose a model in which the physical and nominal dimensions of letter pairs are compared independently of whether subjects use physical (shape task) or nominal (name task) identity as the decision criterion. We attempt to explain the fast-same effect, the preponderance of false-different errors, and the nominal-physical disparity as results of congruent and incongruent outputs of physical and nominal comparison devices that function in both tasks. Subjects performed both tasks with and without response deadlines. The stimuli were presented foveally or unilaterally to one or the other hemisphere. With foveal presentations, the nominal-physical disparity disappeared when congruent and incongruent cells were compared, the fast-same effect occurred only in the shape task, and there was a preponderance of false-different errors only in the name task. Response times and error patterns from centrally presented trials conformed to the predictions of the model. Performance patterns from the lateralized trials conformed only partially. The implications of the data are discussed in the context of several theoretical models of same/different judgments.

Adult↗

PubMed2000

Task switching and the measurement of "switch costs".

The measurement of "switch costs" is held to be of interest because, as is widely believed, they may reflect the control processes that are engaged when subjects switch between two (or more) competing tasks. [In task-switching experiments, the reaction time (RT) switch cost is typically measured as the difference in RT between switch and non-switch (repeat) trials.] In this report we focus on the RT switch costs that remain even after the subject has had some time to prepare for the shift of task, when the switch cost may be approximately asymptotic (so-called residual switch costs). Three experiments are presented. All three experiments used Stroop colour/word, and neutral stimuli. Participants performed the two tasks of word-reading and colour-naming in a regular, double alternation, using the "alternating runs" paradigm (R. D. Rogers & S. Monsell, 1995). The experiments were designed to test the hypothesis that RT switch costs depend on a form of proactive interference (PI) arising from the performance of a prior, competing task. A. Allport, E. A. Styles and S. Hsieh (1994) suggested that these PI effects resulted from "task-set inertia", that is, the persisting activation-suppression of competing task-sets, or competing task-processing pathways. The results confirmed the existence of long-lasting PI from the competing task as a major contributor to switch costs. Non-switch trials, used as the baseline in the measurement of switch costs, were also shown to be strongly affected by similar PI effects. However, task-set inertia was not sufficient to account for these results. The results appeared inconsistent also with all other previous models of task switching. A new hypothesis to explain these between-task interference effects was developed, based on the stimulus-triggered retrieval of competing stimulus-response (S-R) associations, acquired (or strengthened) in earlier trials. Consistent with this retrieval hypothesis, switch costs were shown to depend primarily on the S-R characteristics of the preceding task (the task that was switched from) rather than the upcoming task. Further, the effects of the other, competing task were found to persist over many successive switching trials, affecting switch costs long after the stimulus overlap (and hence the principal S-R competition) between the current tasks had been removed. Switch costs were also found to be affected by recent, item-specific experience with a given stimulus, in either the same or the competing task. Finally, the results showed that switch costs were massively affected by the ratio of the number of prior trials, in response to the same stimuli, that had implemented either the currently intended or the competing S-R mappings. None of these effects are predicted by current models of residual switch costs, which appeal to the differences in control processes assumed to be engaged in switch versus non-switch trials.

Adult↗

PubMed1976

A two-pathway informon theory of conditioning and adaptive pattern recognition.

A neural network theory is proposed which offers an explanation of many of the facts of classical and operant conditioning and adaptive pattern recognition. Interconnected networks of units have been studied and simulated which embody only two rules; firstly, units have inputs from pathways of variable and of fixed conductivity; secondly, the conductivity of a variable pathway is made proportional to the negative of the mutual information function between the signals at its input and output. The signal in a fixed pathway indicates whether the total input to the variable pathways is a member or not of some class. After a learning phase in which the unit, called an informon, receives such labelled inputs, it is able to predict the class of future unlabelled inputs. Such units are stable and their steady state can be calculated.

Conditioning, Psychological↗

PubMed1996

A hypothesis-assessment model of categorical argument strength.

According to the proposed hypothesis-assessment model, the strength of inductive categorical arguments, such as {All Robins Have Substance X therefore All Birds Have Substance X}, is determined by the same factors that affect hypothesis plausibility in the everyday social milieu. The premises of such arguments are viewed as evidence and the conclusion is viewed as a hypothesis. Specifically, the proposed model predicts that the perceived strength of general-conclusion categorical arguments will be a function of (a) the number of premises that instantiate the conclusion; (b) the scope of the conclusion; and (c) the number of accessed alternatives to the conclusion. In Experiment 1, one group rated the strength of individual arguments and another constructed superordinate hypotheses in response to the premise information alone. Most of the variance in perceived argument strength was accounted for by the proposed predictors, R = .94. Experiment 2 employed a new set of arguments and included an additional forced-choice condition in which subjects had to choose the stronger of two arguments. Again, the correlation between predictors and argument strength was high, R = .91, and, all significant forced-choice preferences except one were correctly predicted by the model. The one unpredicted preference suggests the need to include conclusion accessibility as a fourth factor. Also, on a subset of the forced-choice pairs in which no significant preference was observed, two distinct patterns of responding were detected-one predicted and the other unanticipated. Some strengths and limitations of the proposed hypothesis-assessment model are discussed in light of these results.

Adult↗

PubMed2001

Visual information processing using bacteriorhodopsin-based complex LB films.

Image extraction and visual information processing using bacteriorhodopsin (bR)-based bioelectronic devices is presented. Image extraction was achieved using a photoreceptor consisting of bR and spiropyran films. The undesired signals from the photoreceptor were automatically eliminated from the whole signal by spiropyran films acting as an optical noise filter that increases the target signal to an undesired signal ratio. For the information processing, the photoreceptor consisting of bR and lipid films deposited with different configurations was used and the target signals were processed to achieve the pattern recognition. The pattern recognition was based on not only the response variability of bacteriorhodopsin, induced by different film configurations, but also on the initial learning process. The input patterns were predicted by simple calculation with the known signals through the initial learning process.

Bacteriorhodopsins↗

PubMed1998

Decision rules for recognition memory confidence judgments.

According to the standard signal-detection model of recognition memory, confidence judgments for recognition responses are reached in much the same way that old-new decisions are reached (i.e., on the basis of criteria situated along the strength-of-evidence axis). The question investigated here is how the confidence criteria shift when recognition accuracy is manipulated across conditions. Although several theories assume that the old-new decision criterion shifts when recognition accuracy changes, less is known about how the confidence criteria move. An analysis of data previously reported by R. Ratcliff, G. McKoon, and M. Tindall (1994) and some new data reported here suggest that the confidence criteria fan out on the decision axis as d' decreases. This result is qualitatively consistent with the predictions of a likelihood ratio model, although the data did not support the stronger quantitative predictions of this account.

Adult↗

PubMed2000

Is the context specificity of latent inhibition a sufficient explanation of learned irrelevance?

In three experiments, rats were pre-exposed either to uncorrelated presentations of a light and sucrose pellets (group CS/US) or to equivalent presentations of the light and pellets in separate sessions (control). In Experiment 1, subsequent conditioning to the light proceeded more slowly in group CS/US than in the control group, whether this conditioning was excitatory, with the light signalling the delivery of pellets, or inhibitory, with the light signalling their absence. Bonardi and Hall (1996) have argued that this learned irrelevance effect may be reducible to latent inhibition, which would be stronger in group CS/US because they are both pre-exposed and conditioned to the CS in the presence of traces of previous USs occurring in the same session. This analysis implies that group CS/US should have conditioned more rapidly to the CS than controls on the first trial of each session in Experiment 1, but this did not happen. It also implies that the learned irrelevance effect should be reversed if conditioning trials are given at a rate of one per day. Experiments 2 and 3 found no support for this prediction. We conclude that learned irrelevance effects cannot always be reduced to latent inhibition.

Animals↗

PubMed2006

The relationship between absolute and proportion scores of serial order memory: simulation predictions and empirical data.

The capability of a feature model of immediate memory (Naine, 1990; Neath, 2000) to predict and account for a relationship between absolute and proportion scoring of immediate serial recall when memory load is varied (the list-length effect, LLE) is examined. The model correctly predicts the novel finding of an LLE in immediate serial order memory similar to that observed with free recall and previously assumed to be attributable to the long-term memory component of that procedure (Glanzer, 1972). The usefulness of formal models as predictive tools and the continuity between short-term serial order and longer-term item memory are considered.

Attention↗

PubMed2001

Toward a model of false recall: experimental manipulation of encoding context and the collection of verbal reports.

The likelihood of false recall in the Deese-Roediger-McDermott (DRM) paradigm was shown to depend on encoding context in two experiments. When fillers had been preselected to decrease the likelihood of encoding the critical lure's semantic features, false recall was virtually eliminated. However, when the same words were presented rearranged in different presentation orders, levels of false recall that were found in earlier DRM studies (Robinson & Roediger, 1997) were replicated. The role of encoding processes in the DRM paradigm was further explored with additional participants completing the experiment while thinking aloud. During encoding of word lists, participants verbalized semantic elaboration of the critical lure while studying the word lists. A path analysis demonstrated that participants' verbalization of critical lures during encoding reliably predicted their level of false recall.

Adult↗

PubMed2001

Sociocognitive self-regulatory mechanisms governing transgressive behavior.

This longitudinal research examined a structural model of the self-regulatory mechanisms governing transgressive conduct. Perceived academic and self-regulatory efficacy concurrently and longitudinally deterred transgressiveness both directly and by fostering prosocialness and adherence to moral self-sanctions for harmful conduct. The impact of perceived social self-efficacy was mediated through prosocialness. Moral disengagement and prosocialness affected transgressiveness through the mediating influence of irascible affectivity and hostile rumination. Ruminative affectivity, in turn, both concurrently and longitudinally affected transgressiveness. Moral disengagement also contributed independently to variance in transgressiveness over time. This pattern of relations was obtained after controlling for prior transgressiveness. The structural model was replicated across gender and provided a better fit to the data than did several alternative models.

Adolescent↗

PubMed1995

Concept formation vs. logistic regression: predicting death in trauma patients.

This paper discusses two classification models, one based on concept formation and the other using standard logistic regression. The models are first explained in some detail and then evaluated on the same population of trauma patients. The goal of both systems is to predict the outcome of those patients. The results are summarized and explained in terms of differing algorithms of the two models.

Artificial Intelligence↗

PubMed2000

Dopamine-receptor stimulation: biobehavioral and biochemical consequences.

The MPTP monkey is a well-characterized animal model of parkinsonism and provides an exceptional tool for the study of dyskinesias induced by dopamine-like agents. Several such agents have been tested during the past 15 years, and it has been found that the duration of action of these compounds is the most reliable variable with which to predict their dyskinesiogenic profile. It is proposed that L-dopa-induced dyskinesias represent a form of pathological learning caused by chronic pulsatile (nonphysiological) stimulation of dopamine receptors, which activates a cascade of molecular and biochemical events. These events include defective regulation of Fos proteins that belong to the deltaFosB family, increased expression of neuropeptides, and defective GABA- and glutamate-mediated neurotransmission in the output structures of the basal ganglia.

1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine↗

PubMed2004

Selection of patient samples and genes for outcome prediction.

Gene expression profiles with clinical outcome data enable monitoring of disease progression and prediction of patient survival at the molecular level. We present a new computational method for outcome prediction. Our idea is to use an informative subset of original training samples. This subset consists of only short-term survivors who died within a short period and long-term survivors who were still alive after a long follow-up time. These extreme training samples yield a clear platform to identify genes whose expression is related to survival. To find relevant genes, we combine two feature selection methods -- entropy measure and Wilcoxon rank sum test -- so that a set of sharp discriminating features are identified. The selected training samples and genes are then integrated by a support vector machine to build a prediction model, by which each validation sample is assigned a survival/relapse risk score for drawing Kaplan-Meier survival curves. We apply this method to two data sets: diffuse large-B-cell lymphoma (DLBCL) and primary lung adenocarcinoma. In both cases, patients in high and low risk groups stratified by our risk scores are clearly distinguishable. We also compare our risk scores to some clinical factors, such as International Prognostic Index score for DLBCL analysis and tumor stage information for lung adenocarcinoma. Our results indicate that gene expression profiles combined with carefully chosen learning algorithms can predict patient survival for certain diseases.

Biomarkers, Tumor↗

PubMed1995

Molecular mimicry in channel-protein structure.

The approach to probing the sequence-structure relationship of ion-channel proteins using small peptides stems from the abundance of sequence information and the virtual absence of structures at atomic resolution. It is anticipated that model peptides may fold predictably into stable structures and reproduce functional properties of specific proteins. Model peptides are well suited to the application of NMR methods to determine protein structure in a membrane environment or to high-resolution X-ray diffraction analysis. It is timely to ask what we have learned through this strategy and where it may lead in our quest to understand the sequence-structure determinism.

Amino Acid Sequence↗

PubMed2002

Neuroimpairments, activity limitations, and participation restrictions in children with cerebral palsy.

In a representative series of 176 children with cerebral palsy (CP), aged 5 to 8 years, associations were studied between additional neuroimpairments, activity limitations, and participation restrictions in the domains of mobility, education, and social relations as proposed in the International Classification of Functioning Disability and Health (ICF). Learning disability occurred in 40%, epilepsy in 35%, visual impairment in 20%, and infantile hydrocephalus in 9% of the children. Additional neuroimpairments were most frequently seen in children with tetraplegia and dystonic CP and in those with antecedents of brain malformations or severe perinatal compromise. Activity limitations were studied with the Gross Motor Function Classification System (GMFCS) and a system for grading bimanual fine motor function (BFMF) was developed. There was a strong correlation of 0.74 between the GMFCS and BFMF (p<0.001). Learning disability, activity limitations, and participation restrictions were all clinically strongly associated with each other (p<0.001). Restriction in mobility was best predicted by the GMFCS, learning disability, and the BFMF; in education by learning disability and the GFMCS; and in social relations by learning disability, the GMFCS, and BFMF. Motor function and learning disability were important predictors for participation restrictions in children with CP. The ICF has the capacity to be a model to help plan interventions for specific functional goals and to ascertain the child's participation in society.

Activities of Daily Living↗

PubMed2004

Optimality principles in sensorimotor control.

The sensorimotor system is a product of evolution, development, learning and adaptation-which work on different time scales to improve behavioral performance. Consequently, many theories of motor function are based on 'optimal performance': they quantify task goals as cost functions, and apply the sophisticated tools of optimal control theory to obtain detailed behavioral predictions. The resulting models, although not without limitations, have explained more empirical phenomena than any other class. Traditional emphasis has been on optimizing desired movement trajectories while ignoring sensory feedback. Recent work has redefined optimality in terms of feedback control laws, and focused on the mechanisms that generate behavior online. This approach has allowed researchers to fit previously unrelated concepts and observations into what may become a unified theoretical framework for interpreting motor function. At the heart of the framework is the relationship between high-level goals, and the real-time sensorimotor control strategies most suitable for accomplishing those goals.

Adaptation, Physiological↗

PubMed2004

Data mining in bioinformatics using Weka.

UNLABELLED: The Weka machine learning workbench provides a general-purpose environment for automatic classification, regression, clustering and feature selection-common data mining problems in bioinformatics research. It contains an extensive collection of machine learning algorithms and data pre-processing methods complemented by graphical user interfaces for data exploration and the experimental comparison of different machine learning techniques on the same problem. Weka can process data given in the form of a single relational table. Its main objectives are to (a) assist users in extracting useful information from data and (b) enable them to easily identify a suitable algorithm for generating an accurate predictive model from it. AVAILABILITY: http://www.cs.waikato.ac.nz/ml/weka.

Algorithms↗