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 631 records · Page 35Linked to original sources

PubMed1983

Learned helplessness: the role of individual differences in learned resourcefulness.

The term 'learned resourcefulness' refers to an acquired repertoire of behaviours and skills by which a person self-regulates internal events (such as emotions, pain, and cognitions) that interfere with the smooth execution of a target behaviour. Sixty undergraduate students were rated as either high resourceful (HR) or low resourceful (LR) according to their scores on Rosenbaum's Self-Control Schedule. Subjects were then pre-treated with inescapable, escapable or control aversive tone followed by anagram solution testing. As hypothesized the learned helplessness phenomenon, the interference with new learning following inescapable aversive events, appeared only in LR subjects and not in HR subjects. No relationship was found between subjects' causal attributions for their performance on the noise task and their subsequent performance on the anagrams as would be predicted from the attributional part of the recent reformulation of the learned helplessness model. It was concluded that subjects' general repertoire of self-control skills, and their general expectations for self-efficacy, might be at least as important in explaining the generalization of helplessness from the training task to the test task as the kinds of causal attributions subjects make for their performance on the training task.

Adaptation, Psychological↗

PubMed2003

Isotropic sequence order learning.

In this article, we present an isotropic unsupervised algorithm for temporal sequence learning. No special reward signal is used such that all inputs are completely isotropic. All input signals are bandpass filtered before converging onto a linear output neuron. All synaptic weights change according to the correlation of bandpass-filtered inputs with the derivative of the output. We investigate the algorithm in an open- and a closed-loop condition, the latter being defined by embedding the learning system into a behavioral feedback loop. In the open-loop condition, we find that the linear structure of the algorithm allows analytically calculating the shape of the weight change, which is strictly heterosynaptic and follows the shape of the weight change curves found in spike-time-dependent plasticity. Furthermore, we show that synaptic weights stabilize automatically when no more temporal differences exist between the inputs without additional normalizing measures. In the second part of this study, the algorithm is is placed in an environment that leads to closed sensor-motor loop. To this end, a robot is programmed with a prewired retraction reflex reaction in response to collisions. Through isotropic sequence order (ISO) learning, the robot achieves collision avoidance by learning the correlation between his early range-finder signals and the later occurring collision signal. Synaptic weights stabilize at the end of learning as theoretically predicted. Finally, we discuss the relation of ISO learning with other drive reinforcement models and with the commonly used temporal difference learning algorithm. This study is followed up by a mathematical analysis of the closed-loop situation in the companion article in this issue, "ISO Learning Approximates a Solution to the Inverse-Controller Problem in an Unsupervised Behavioral Paradigm" (pp. 865-884).

Algorithms↗

PubMed2004

In silico estimation of DMSO solubility of organic compounds for bioscreening.

Solubility of organic compounds in DMSO is an important issue for commercial and academic organizations handling large compound collections or performing biological screening. In particular, solubility data are critical for the optimization of storage conditions and for the selection of compounds for bioscreening compatible with the assay protocol. Solubility is largely determined by the solvation energy and the crystal disruption energy, and these molecular phenomena should be assessed in structure-solubility correlation studies. The authors summarize our long-term experimental observations and theoretical studies of physicochemical determinants of DMSO solubility of organic substances. They compiled a comprehensive reference database of proprietary data on compound solubility (55,277 compounds with good DMSO solubility and 10,223 compounds with poor DMSO solubility), calculated specific molecular descriptors (topological, electromagnetic, charge, and lipophilicity parameters), and applied an advanced machine-learning approach for training neural networks to address the solubility. Both supervised (feed-forward, back-propagated neural networks) and unsupervised (Kohonen neural networks) learning methods were used. The resulting neural network models were validated by successfully predicting DMSO solubility of compounds in independent test selections.

Dimethyl Sulfoxide↗

PubMed1987

Three accounts of the learned helplessness effect.

Two experiments tested predictions drawn from test anxiety theory, learned helplessness theory, and Wortman and Brehm's (1975) integration of helplessness and reactance theories. Experiment 1 demonstrated that performance deficits predicted by learned helplessness do not rely on experimenter-induced failure. It also showed such deficits to be unrelated either to negative affect following exposure to pretreatment or to causal attributions about pretreatment task performance. Experiment 2 showed that experience of uncontrollability need not result in impaired performance, because failure on an unimportant task did not produce the deficits predicted by learned helplessness theory. This result provides qualified support for the integrative model. Finally, because the subjective measures used in Experiment 2 were not consistent with performance measures, the reliability of self-reports is questioned.

Achievement↗

PubMed2007

Predicting costs over time using Bayesian Markov chain Monte Carlo methods: an application to early inflammatory polyarthritis.

This article focuses on the modelling and prediction of costs due to disease accrued over time, to inform the planning of future services and budgets. It is well documented that the modelling of cost data is often problematic due to the distribution of such data; for example, strongly right skewed with a significant percentage of zero-cost observations. An additional problem associated with modelling costs over time is that cost observations measured on the same individual at different time points will usually be correlated. In this study we compare the performance of four different multilevel/hierarchical models (which allow for both the within-subject and between-subject variability) for analysing healthcare costs in a cohort of individuals with early inflammatory polyarthritis (IP) who were followed-up annually over a 5-year time period from 1990/1991. The hierarchical models fitted included linear regression models and two-part models with log-transformed costs, and two-part model with gamma regression and a log link. The cohort was split into a learning sample, to fit the different models, and a test sample to assess the predictive ability of these models. To obtain predicted costs on the original cost scale (rather than the log-cost scale) two different retransformation factors were applied. All analyses were carried out using Bayesian Markov chain Monte Carlo (MCMC) simulation methods.

Adult↗

PubMed2005

Formal learning theory dissociates brain regions with different temporal integration.

Learning can be characterized as the extraction of reliable predictions about stimulus occurrences from past experience. In two experiments, we investigated the interval of temporal integration of previous learning trials in different brain regions using implicit and explicit Pavlovian fear conditioning with a dynamically changing reinforcement regime in an experimental setting. With formal learning theory (the Rescorla-Wagner model), temporal integration is characterized by the learning rate. Using fMRI and this theoretical framework, we are able to distinguish between learning-related brain regions that show long temporal integration (e.g., amygdala) and higher perceptual regions that integrate only over a short period of time (e.g., fusiform face area, parahippocampal place area). This approach allows for the investigation of learning-related changes in brain activation, as it can dissociate brain areas that differ with respect to their integration of past learning experiences by either computing long-term outcome predictions or instantaneous reinforcement expectancies.

Adult↗

PubMed2004

Population of linear experts: knowledge partitioning and function learning.

Knowledge partitioning is a theoretical construct holding that knowledge is not always integrated and homogeneous but may be separated into independent parcels containing mutually contradictory information. Knowledge partitioning has been observed in research on expertise, categorization, and function learning. This article presents a theory of function learning (the population of linear experts model--POLE) that assumes people partition their knowledge whenever they are presented with a complex task. The authors show that POLE is a general model of function learning that accommodates both benchmark results and recent data on knowledge partitioning. POLE also makes the counterintuitive prediction that a person's distribution of responses to repeated test stimuli should be multimodal. The authors report 3 experiments that support this prediction.

Analysis of Variance↗

PubMed1984

Attributional style: degree of depression, respondent's sex, and nature of the attributional event.

Attributional responses made to both good and bad events by male and female Ss (N = 117) differing in degree of clinical depression were examined in light of the reformulated learned helplessness model of depression. Data comparison with control Ss provided support for the model's prediction. Clinically depressed Ss assumed more personal responsibility than control Ss for their life outcomes, especially the negative ones. While significant main effects were demonstrated in the clinical sample for all three variables, the significant two and three-way interaction effects revealed a complex relationship between causal attribution and depression.

Adaptation, Psychological↗

PubMed2025

Genetic mapping and predictive modeling of paralog synthetic lethality.

Paralogs are abundant in the human genome and thought to be a primary source of synthetic lethality, yet the vast paralogome remains largely uncharacterized. A digenic screen of 36,648 paralogous pairs in the human genome revealed that synthetic lethalities were infrequent and varied in penetrance in different tumor backgrounds. We hypothesized that the variable penetrance of synthetic lethalities resulted from complex polygenic interactions with different cellular contexts. A machine learning classifier of a subset of paralog pairs tested across 49 cancer models revealed that endogenous perturbations in related pathways predicted paralog synthetic lethality. Further, predictive modeling of paralog synthetic lethality showed that the strength of synthetic lethal interactions was largely due to the overlap and essentiality of the protein-protein interaction networks shared by the paralog pairs. Collectively, this study tested 36,648 digenic paralog interactions and delineated the key feature classes that underlie the heterogeneity of paralog synthetic lethalities.

Humans↗

PubMed2025

A machine learning model and identification of immune infiltration for chronic obstructive pulmonary disease based on disulfidptosis-related genes.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is a chronic and progressive lung disease. Disulfidptosis-related genes (DRGs) may be involved in the pathogenesis of COPD. From the perspective of predictive, preventive, and personalized medicine (PPPM), clarifying the role of disulfidptosis in the development of COPD could provide a opportunity for primary prediction, targeted prevention, and personalized treatment of the disease. METHODS: We analyzed the expression profiles of DRGs and immune cell infiltration in COPD patients by using the GSE38974 dataset. According to the DRGs, molecular clusters and related immune cell infiltration levels were explored in individuals with COPD. Next, co-expression modules and cluster-specific differentially expressed genes were identified by the Weighted Gene Co-expression Network Analysis (WGCNA). Comparing the performance of the random forest (RF), support vector machine (SVM), generalized linear model (GLM), and eXtreme Gradient Boosting (XGB), we constructed the ptimal machine learning model. RESULTS: DE-DRGs, differential immune cells and two clusters were identified. Notable difference in DRGs, immune cell populations, biological processes, and pathway behaviors were noted among the two clusters. Besides, significant differences in DRGs, immune cells, biological functions, and pathway activities were observed between the two clusters.A nomogram was created to aid in the practical application of clinical procedures. The SVM model achieved the best results in differentiating COPD patients across various clusters. Following that, we identified the top five genes as predictor genes via SVM model. These five genes related to the model were strongly linked to traits of the individuals with COPD. CONCLUSION: Our study demonstrated the relationship between disulfidptosis and COPD and established an optimal machine-learning model to evaluate the subtypes and traits of COPD. DRGs serve as a target for future predictive diagnostics, targeted prevention, and individualized therapy in COPD, facilitating the transition from reactive medical services to PPPM in the management of the disease.

Pulmonary Disease, Chronic Obstructive↗

PubMed2026

Transcriptome-based high-frequency recurrence index predicts frequent recurrence in non-muscle-invasive bladder cancer after Bacillus Calmette-Guérin therapy.

BACKGROUND: High-frequency recurrence (HfR, ≥ 2 recurrences) in non-muscle-invasive bladder cancer (NMIBC) poses a significant clinical burden. Current risk models, such as the European Organization for Research and Treatment of Cancer (EORTC), the European Association of Urology (EAU), and the UROMOL classification, offer limited predictive accuracy for identifying patients at risk for frequent recurrence despite appropriate treatment. METHODS: A 75-gene high-frequency recurrence index (HfRI) was constructed by selecting recurrence-associated genes using differential expression and Cox regression analyses. The HfRI was computed as a weighted sum of normalized gene expression values. The model was trained on a discovery cohort and validated in multiple cohorts (n = 1379) using machine-learning approaches. Clinical relevance was assessed using recurrence-free survival (RFS) and Cox models, and predictive performance was compared with that of the EORTC, EAU, and UROMOL classifications using the area under the curve (AUC) and the concordance index (c-index). RESULTS: The HfRI robustly stratified patients into high-risk and low-risk groups across six independent NMIBC cohorts. Patients classified as HfRI-high had a significantly greater likelihood of experiencing ≥ 2 recurrences (χ2, p = 0.001) and showed markedly reduced RFS (log-rank test, p < 0.001). The adverse prognostic effect of the HfRI persisted even among patients treated with BCG therapy (log-rank test, p = 0.02). Multivariate analysis revealed that the HfRI was an independent predictor of HfR (HR = 2.82, 95% CI = 1.89-4.20, p < 0.001). Compared with established clinical risk classifiers, the HfRI demonstrated superior predictive performance (AUC = 0.736, c-index = 0.673) in terms of the EORTC (AUC = 0.594), EAU (AUC = 0.557) risk groups, and UROMOL2021 (AUC = 0.596) classification. Pathway analysis revealed that HfRI-high tumors were characterized by upregulation of cell cycle progression and DNA replication pathways, accompanied by suppression of immune signaling pathways. These biological features provide a mechanistic explanation for the reduced responsiveness to intravesical BCG therapy, underscoring the role of HfRI not only as a predictor of recurrence risk but also as a biomarker capable of identifying patients unlikely to benefit from standard BCG treatment. CONCLUSIONS: HfRI represents a robust, transcriptome-based tool for predicting frequent recurrence in NMIBC patients. The HfRI supports earlier identification of patients at risk of high-frequency recurrence, thereby supporting personalized treatment strategies.

Humans↗

PubMed1998

Identification of 'genotoxic' and 'non-genotoxic' alerts for cancer in mice: the carcinogenic potency database.

A set of chemicals tested for carcinogenicity in mice that have been analyzed by Gold et al. [L.S. Gold, C.B. Sawyer, R. Magaw, G.M. Backman, M. deVeciana, R. Levinson, N.K. Hooper, W.R. Havender, L. Bernstein, R. Peto, M.C. Pike, B.N. Ames, Environ. Health Perspect. 58 (1984) 9-319; L.S. Gold, M. deVeciana, G.M. Backman, M. Lopipero, M. Smith, R. Blumenthal, R. Levinson, L. Bernstein, B.N. Ames, Environ. Health Perspect. 67 (1986) 161-200; L.S. Gold, T.H. Slone, G.M. Backman, R. Magaw, M. DaCosta, P. Lopipero, M. Blumenthal, B.N. Ames, Environ. Health Perspect. 74 (1987) 237-329; L.S. Gold, T.H. Slone, G.M. Backman, S. Eisenberg, M. DaCosta, M. Wong, N.B. Manley, L. Rohrbach, B.N. Ames, Environ. Health Perspect. 84 (1990) 215-286; L.S. Gold, N.B. Manley, T.H. Slone, T.H. Garfinkle, L. Rohrbach, B.N. Ames, Environ. Health Perspect. 100 (1993) 65-135] in the first five plots of the carcinogenic potency database (CPDB) was subjected to CASE/MULTICASE analyses. Briefly, CASE/MULTICASE is a computer-automated structure evaluation system that is capable of identifying structural features of chemicals associated with a specified biological activity (e.g., carcinogenicity or mutagenicity). These features are then incorporated into a structure-activity relationship (SAR) model for the analyzed database. The mouse CPDB used in this study consists of 627 chemicals, 289 of which are carcinogens, 11 marginal or weak carcinogens (i.e., chemicals requiring high doses to induce cancer) and 327 non-carcinogens. In an internal prediction analysis where the CASE/MULTICASE SAR model was used to predict the carcinogenicity of chemicals used to create the model, a concordance between experimental and predicted results of 96% was obtained. This indicates that the model is able to satisfactorily explain the chemicals in the learning set. In a drop-one cross-validation study where chemicals were removed one at a time and the remaining n - 1 chemicals were used in an iterative method to create a model to predict the removed chemical, CASE/MULTICASE was able to achieve a concordance between experimental and predicted results of 70%. Using a modified validation process designed to investigate the predictivity of a more focused SAR model, the system achieved a 78% concordance between experimental and predicted results. Among the major biophores identified by CASE/MULTICASE associated with cancer causation in mice several are derived from electrophilic or potentially electrophilic compounds (e.g., hydrazines, N-mustards, N-nitrosamines, aromatic amines, reactive halogens, and quinones). Other biophores however are derived from chemicals seemingly devoid of actual or potential DNA-reactivity and as such may represent structural feature of non-genotoxic carcinogens.

Animals↗

PubMed2004

A new model for prediction of the age of onset and penetrance for Huntington's disease based on CAG length.

Huntington's disease (HD) is a neurodegenerative disorder caused by an unstable CAG repeat. For patients at risk, participating in predictive testing and learning of having CAG expansion, a major unanswered question shifts from "Will I get HD?" to "When will it manifest?" Using the largest cohort of HD patients analyzed to date (2913 individuals from 40 centers worldwide), we developed a parametric survival model based on CAG repeat length to predict the probability of neurological disease onset (based on motor neurological symptoms rather than psychiatric onset) at different ages for individual patients. We provide estimated probabilities of onset associated with CAG repeats between 36 and 56 for individuals of any age with narrow confidence intervals. For example, our model predicts a 91% chance that a 40-year-old individual with 42 repeats will have onset by the age of 65, with a 95% confidence interval from 90 to 93%. This model also defines the variability in HD onset that is not attributable to CAG length and provides information concerning CAG-related penetrance rates.

Age of Onset↗

PubMed1985

Learned helplessness and learned resourcefulness: effects of noncontingent success and failure on individuals differing in self-control skills.

We designed two experiments to investigate the role of self-control processes in learned-helplessness studies by assessing the differential reactions to uncontrollability of subjects who presumably had either a rich (high resourceful, or HR) or poor (low resourceful, or LR) repertoire of self-control skills. HR and LR subjects received noncontingent success feedback, failure feedback, or no feedback on a task that ostensibly assessed "therapeutic abilities." Subjects were subsequently tested on insolvable puzzles (Experiment 1) or on solvable anagrams (Experiment 2). According to Kanfer and Hagerman's (1981) self-regulation model, self-regulatory activities are evoked primarily in situations in which subjects are faced with repeated failure. Hence we predicted that individual differences in self-control would influence performance on the insolvable puzzles and not anagram performance after exposure to noncontingent failure. This prediction was confirmed: Only the performance of LR subjects on the insolvable puzzles was debilitated by the helplessness induction, whereas HR and LR subjects showed equal helplessness-induced deficits on the anagrams. The latter finding was interpreted in terms of the learned-helplessness model without the mediating effects of self-regulatory processes. As predicted from the self-control model, HR subjects more frequently checked statements indicating positive self-evaluations and task-oriented thoughts and less frequently checked negative self-evaluations than did LR subjects during exposure to uncontrollability in both experiments. We concluded that the self-control model accounts best for subjects' self-reactions during exposure to uncontrollability or failure, whereas the learned-helplessness model accounts for the generalization of helplessness from uncontrollable situations to controllable ones.

Environment↗

PubMed1999

Working memory, executive processes and the effects of alcohol on Go/No-Go learning: testing a model of behavioral regulation and impulsivity.

RATIONALE: Impulsivity is associated with increased risk for alcoholism. Alcohol also may increase impulsive behavior, although little is known about the processes underlying this effect. OBJECTIVES: This study tested a model proposing that the executive processes of working memory (WM) and conditional associative learning (CAL) modulate behavioral inhibition. Subjects had either a positive (FHP) or a negative (FHN) family history of alcoholism. Hypotheses were that alcohol would increase Go/No-Go impulsive responding but only in subjects with low working memory capacity (low-WM), low-CAL ability, or FHP for alcoholism. The model also predicted that WM and CAL modulate inhibitory responses to contingency reversal on a Go/No-Go task. METHODS: A Go/No-Go learning task with a midway contingency reversal was administered to 71 FHP and 78 FHN subjects when sober and after drinking one of two moderate doses of alcohol. WM (digits backward) and CAL (conditional spatial association task) were also assessed when sober. RESULTS: Alcohol resulted in more false alarms but only in low-WM subjects. Both WM and CAL modulated learning to inhibit behavior after contingency reversal, suggesting separate modulation mechanisms for WM and CAL. Subjects with low- capacity WM and subjects with low-capacity CAL ability had more difficulty learning response inhibition after contingency reversal. FHPs and FHNs did not differ in their response to alcohol. CONCLUSIONS: The results support our model of the modulatory role of WM and CAL in the ongoing regulation of behavioral inhibitory systems. The results also suggest that individuals with low capacity WM are more susceptible to alcohol's effect of increasing impulsive behavior, suggesting that alcohol reduces the capacity of working memory to modulate response inhibition.

Adolescent↗

PubMed2006

A compression-based approach for coding sequences identification. I. Application to prokaryotic genomes.

Most of the gene prediction algorithms for prokaryotes are based on Hidden Markov Models or similar machine-learning approaches, which imply the optimization of a high number of parameters. The present paper presents a novel method for the classification of coding and non-coding regions in prokaryotic genomes, based on a suitably defined compression index of a DNA sequence. The main features of this new method are the non-parametric logic and the costruction of a dictionary of words extracted from the sequences. These dictionaries can be very useful to perform further analyses on the genomic sequences themselves. The proposed approach has been applied on some prokaryotic complete genomes, obtaining optimal scores of correctly recognized coding and non-coding regions. Several false-positive and false-negative cases have been investigated in detail, which have revealed that this approach can fail in the presence of highly structured coding regions (e.g., genes coding for modular proteins) or quasi-random non-coding regions (e.g., regions hosting non-functional fragments of copies of functional genes; regions hosting promoters or other protein-binding sequences). We perform an overall comparison with other gene-finder software, since at this step we are not interested in building another gene-finder system, but only in exploring the possibility of the suggested approach.

Algorithms↗

PubMed2004

Learning multiple visuomotor transformations: adaptation and context-dependent recall.

Recent motor control theories suggest that the brain uses internal models to plan and control accurate movements. An internal model is thought to represent how the biomechanics of the arm interacting with the outside world would respond to a motor command; therefore it can be seen as a predictive model of the reafference that helps the system plan ahead. Moreover, adaptation studies show that humans can learn multiple internal models. It is not clear, however, whether and how contextual cues are used to switch among competing internal models, which are required to compensate for altered environments. To investigate this question, we asked healthy participants to perform center-out pointing movements under normal and distorted visual feedback (0 degrees , 30 degrees counterclockwise, and 60 degrees clockwise rotation of hand-screen cursor relationships) conditions. The results suggest that humans can learn multiple environments simultaneously and can use contextual cues to facilitate adaptation and to recall the appropriate internal model of the visuomotor transformation.

Adaptation, Physiological↗

PubMedDate not supplied

Boosting: an ensemble learning tool for compound classification and QSAR modeling.

A classification and regression tool, J. H. Friedman's Stochastic Gradient Boosting (SGB), is applied to predicting a compound's quantitative or categorical biological activity based on a quantitative description of the compound's molecular structure. Stochastic Gradient Boosting is a procedure for building a sequence of models, for instance regression trees (as in this paper), whose outputs are combined to form a predicted quantity, either an estimate of the biological activity, or a class label to which a molecule belongs. In particular, the SGB procedure builds a model in a stage-wise manner by fitting each tree to the gradient of a loss function: e.g., squared error for regression and binomial log-likelihood for classification. The values of the gradient are computed for each sample in the training set, but only a random sample of these gradients is used at each stage. (Friedman showed that the well-known boosting algorithm, AdaBoost of Freund and Schapire, could be considered as a particular case of SGB.) The SGB method is used to analyze 10 cheminformatics data sets, most of which are publicly available. The results show that SGB's performance is comparable to that of Random Forest, another ensemble learning method, and are generally competitive with or superior to those of other QSAR methods. The use of SGB's variable importance with partial dependence plots for model interpretation is also illustrated.

ATP Binding Cassette Transporter, Subfamily B, Mem↗