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 1,405 records · Page 78Linked to original sources

PubMed2005

Integrating seasonal climate prediction and agricultural models for insights into agricultural practice.

Interest in integrating crop simulation models with dynamic seasonal climate forecast models is expanding in response to a perceived opportunity to add value to seasonal climate forecasts for agriculture. Integrated modelling may help to address some obstacles to effective agricultural use of climate information. First, modelling can address the mismatch between farmers' needs and available operational forecasts. Probabilistic crop yield forecasts are directly relevant to farmers' livelihood decisions and, at a different scale, to early warning and market applications. Second, credible ex ante evidence of livelihood benefits, using integrated climate-crop-economic modelling in a value-of-information framework, may assist in the challenge of obtaining institutional, financial and political support; and inform targeting for greatest benefit. Third, integrated modelling can reduce the risk and learning time associated with adaptation and adoption, and related uncertainty on the part of advisors and advocates. It can provide insights to advisors, and enhance site-specific interpretation of recommendations when driven by spatial data. Model-based 'discussion support systems' contribute to learning and farmer-researcher dialogue. Integrated climate-crop modelling may play a genuine, but limited role in efforts to support climate risk management in agriculture, but only if they are used appropriately, with understanding of their capabilities and limitations, and with cautious evaluation of model predictions and of the insights that arises from model-based decision analysis.

Climate↗

PubMedDate not supplied

Contributions of memory circuits to language: the declarative/procedural model.

The structure of the brain and the nature of evolution suggest that, despite its uniqueness, language likely depends on brain systems that also subserve other functions. The declarative/procedural (DP) model claims that the mental lexicon of memorized word-specific knowledge depends on the largely temporal-lobe substrates of declarative memory, which underlies the storage and use of knowledge of facts and events. The mental grammar, which subserves the rule-governed combination of lexical items into complex representations, depends on a distinct neural system. This system, which is composed of a network of specific frontal, basal-ganglia, parietal and cerebellar structures, underlies procedural memory, which supports the learning and execution of motor and cognitive skills, especially those involving sequences. The functions of the two brain systems, together with their anatomical, physiological and biochemical substrates, lead to specific claims and predictions regarding their roles in language. These predictions are compared with those of other neurocognitive models of language. Empirical evidence is presented from neuroimaging studies of normal language processing, and from developmental and adult-onset disorders. It is argued that this evidence supports the DP model. It is additionally proposed that "language" disorders, such as specific language impairment and non-fluent and fluent aphasia, may be profitably viewed as impairments primarily affecting one or the other brain system. Overall, the data suggest a new neurocognitive framework for the study of lexicon and grammar.

Humans↗

PubMedDate not supplied

Towards a neural model of timing.

This paper describes a neural model of interval timing, which reproduces the duration discrimination experiments of Wearden, J.H., 1992, J. Exp. Psychol. 18, 134-144. The model comprises three layers of neural units. The units in the first layer represent clusters of neurons with probabilistic internal feedback that maintains self-sustained (short-term memory) activity for a random time. The unit in the second layer is a spiking neuron that fires as long as a sufficient number of input clusters are active. The unit in the third layer detects the offset of firing in the previous layer by producing a short burst of spikes. Analysis and simulation of the model shows spikes produced at random times with a distribution determined by the number of units in the first layer, their survival time constant, and the threshold of the unit in layer 2. Interval times can be learned with any of these parameters but lead to different Weber law relations. A variable threshold in layer 2 predicts S-shaped Weber curves, a variable number of units in layer 1 leads to a saturation of the Weber curve (decreasing Weber fraction) and a variable time constant in layer 1 causes a linear Weber curve.

Models, Neurological↗

PubMed2025

Assessing data size requirements for training generalizable sequence-based TCR specificity models via pan-allelic MHC-I point-mutation ligandome evaluation.

Rapid identification of T cell receptors (TCRs) that specifically bind patient-unique neoepitopes is a critical challenge for personalized TCR-based therapies in oncology. Due to enormous diversity of both TCR and neoepitope repertoires, a machine learning predictor of TCR-pMHC specificity for personalized therapy must generalize to TCRs and epitopes not seen in the training data. We estimate the necessary size of such training data. We first confirm that published models fail to generalize beyond a single-residue dissimilarity to the epitope training set distribution. We then impute the point-mutation ligandome across the 34 most prevalent human MHC alleles and represent it as a graph based on our established dissimilarity cutoff. By finding the dominating set of this graph, we estimate that between one and 100 million epitopes are required to train a generalizable sequence-based TCR specificity prediction model-1000 times the size of current public data.

Humans↗

PubMed2024

Blood-based DNA methylation and exposure risk scores predict PTSD with high accuracy in military and civilian cohorts.

BACKGROUND: Incorporating genomic data into risk prediction has become an increasingly popular approach for rapid identification of individuals most at risk for complex disorders such as PTSD. Our goal was to develop and validate Methylation Risk Scores (MRS) using machine learning to distinguish individuals who have PTSD from those who do not. METHODS: Elastic Net was used to develop three risk score models using a discovery dataset (n = 1226; 314 cases, 912 controls) comprised of 5 diverse cohorts with available blood-derived DNA methylation (DNAm) measured on the Illumina Epic BeadChip. The first risk score, exposure and methylation risk score (eMRS) used cumulative and childhood trauma exposure and DNAm variables; the second, methylation-only risk score (MoRS) was based solely on DNAm data; the third, methylation-only risk scores with adjusted exposure variables (MoRSAE) utilized DNAm data adjusted for the two exposure variables. The potential of these risk scores to predict future PTSD based on pre-deployment data was also assessed. External validation of risk scores was conducted in four independent cohorts. RESULTS: The eMRS model showed the highest accuracy (92%), precision (91%), recall (87%), and f1-score (89%) in classifying PTSD using 3730 features. While still highly accurate, the MoRS (accuracy = 89%) using 3728 features and MoRSAE (accuracy = 84%) using 4150 features showed a decline in classification power. eMRS significantly predicted PTSD in one of the four independent cohorts, the BEAR cohort (beta = 0.6839, p=0.006), but not in the remaining three cohorts. Pre-deployment risk scores from all models (eMRS, beta = 1.92; MoRS, beta = 1.99 and MoRSAE, beta = 1.77) displayed a significant (p < 0.001) predictive power for post-deployment PTSD. CONCLUSION: The inclusion of exposure variables adds to the predictive power of MRS. Classification-based MRS may be useful in predicting risk of future PTSD in populations with anticipated trauma exposure. As more data become available, including additional molecular, environmental, and psychosocial factors in these scores may enhance their accuracy in predicting PTSD and, relatedly, improve their performance in independent cohorts.

Humans↗

PubMed2004

Approaches to learning and studying in medical students: validation of a revised inventory and its relation to student characteristics and performance.

INTRODUCTION: Inventories to quantify approaches to studying try to determine how students approach academic tasks. Medical curricula usually aim to promote a deep approach to studying, which is associated with academic success and which may predict desirable traits postqualification. AIMS: This study aimed to validate a revised Approaches to Learning and Studying Inventory (ALSI) in medical students and to explore its relation to student characteristics and performance. METHODS: Confirmatory factor analysis was used to validate the reported constructs in a sample of 128 Year 1 medical students. Models were developed to investigate the effect of age, graduate status and gender, and the relationships between approaches to studying and assessment outcomes. RESULTS: The ALSI performed as anticipated in this population, thus validating its use in our sample, but a 4-factor solution had a better fit than the reported 5-factor one. Medical students scored highly on deep approach compared with other students in higher education. Graduate status and gender had significant effects on approach to studying and a deep approach was associated with higher academic scores. CONCLUSIONS: The ALSI is valid for use in medical students and can uncover interesting relationships between approaches to studying and student characteristics. In addition, the ALSI has potential as a tool to predict student success, both academically and beyond qualification.

Age Factors↗

PubMed2005

The micro-module learning tests: work-sample assessments of responsiveness to skills training.

While outcome of psychiatric rehabilitation has been successfully predicted by cognitive tests, efforts to design a measure to assess responsiveness to rehabilitation have been lacking. In this report, we describe the rationale for and development of a face-valid measure of responsivity to the three core components of skills training: responsiveness to verbal instruction, ability to learn from viewing the behavior of a model, and ability to demonstrate skills observed during a subsequent role-play. Seven alternate forms of the new measure, called the Micro-Module Learning Test (MMLT), demonstrated adequate internal consistency and alternate-form reliability. We also present results from four studies in which the MMLT was used to collect normative data as well as data on relationships with symptoms, cognitive tests, and treatment outcome. Results indicate that the MMLT is associated with cognitive factors found to predict treatment outcome in prior studies (e.g., verbal memory and fluency), as well as lesser investigated functions such as theory of mind ability. In addition, MMLT scores were correlated negatively with psychotic disorganization and positively with performance during a full-length skills training group. The MMLT appears to be a reliable and valid measure for rapidly assessing responsiveness to skills training procedures.

Adult↗

PubMed2003

A gain-field encoding of limb position and velocity in the internal model of arm dynamics.

Adaptability of reaching movements depends on a computation in the brain that transforms sensory cues, such as those that indicate the position and velocity of the arm, into motor commands. Theoretical consideration shows that the encoding properties of neural elements implementing this transformation dictate how errors should generalize from one limb position and velocity to another. To estimate how sensory cues are encoded by these neural elements, we designed experiments that quantified spatial generalization in environments where forces depended on both position and velocity of the limb. The patterns of error generalization suggest that the neural elements that compute the transformation encode limb position and velocity in intrinsic coordinates via a gain-field; i.e., the elements have directionally dependent tuning that is modulated monotonically with limb position. The gain-field encoding makes the counterintuitive prediction of hypergeneralization: there should be growing extrapolation beyond the trained workspace. Furthermore, nonmonotonic force patterns should be more difficult to learn than monotonic ones. We confirmed these predictions experimentally.

Adult↗

PubMed2005

Word length effects in Hebrew.

Numerous lateralization studies have reported that word length has a stronger effect in the left visual field (LVF) than in the right visual field (RVF) for right-handed people due to hemispheric asymmetry for language processing. Alternatively, early perceptual learning theory argued that the length effects might depend on the frequency of having read words at various lengths displayed at different retinal locations. The two alternatives were tested with right-handers participants who were native speakers of Hebrew which is read from right to left, that is Hebrew readers have a different perceptual experience than English readers. We found the predicted interaction between word length and hemifield; however, longer latencies to longer letter strings were found at both visual fields. We argue that these results are best accounted by the SERIOL model of letter-position encoding.

Adult↗

PubMed2005

Sequence-structure-function relationships of a tRNA (m7G46) methyltransferase studied by homology modeling and site-directed mutagenesis.

The Escherichia coli TrmB protein and its Saccharomyces cerevisiae ortholog Trm8p catalyze the S-adenosyl-L-methionine-dependent formation of 7-methylguanosine at position 46 (m7G46) in tRNA. To learn more about the sequence-structure-function relationships of these enzymes we carried out a thorough bioinformatics analysis of the tRNA:m7G methyltransferase (MTase) family to predict sequence regions and individual amino acid residues that may be important for the interactions between the MTase and the tRNA substrate, in particular the target guanosine 46. We used site-directed mutagenesis to construct a series of alanine substitutions and tested the activity of the mutants to elucidate the catalytic and tRNA-recognition mechanism of TrmB. The functional analysis of the mutants, together with the homology model of the TrmB structure and the results of the phylogenetic analysis, revealed the crucial residues for the formation of the substrate-binding site and the catalytic center in tRNA:m7G MTases.

Amino Acid Sequence↗

PubMed2000

The effect of intracortical competition on the formation of topographic maps in models of Hebbian learning.

Correlation-based learning (CBL) models and self-organizing maps (SOM) are two classes of Hebbian models that have both been proposed to explain the activity-driven formation of cortical maps. Both models differ significantly in the way lateral cortical interactions are treated, leading to different predictions for the formation of receptive fields. The linear CBL models predict that receptive field profiles are determined by the average values and the spatial correlations of the second order of the afferent activity patterns, whereas SOM models map stimulus features. Here, we investigate a class of models which are characterized by a variable degree of lateral competition and which have the CBL and SOM models as limit cases. We show that there exists a critical value for intracortical competition below which the model exhibits CBL properties and above which feature mapping sets in. The class of models is then analyzed with respect to the formation of topographic maps between two layers of neurons. For Gaussian input stimuli we find that localized receptive fields and topographic maps emerge above the critical value for intracortical competition, and we calculate this value as a function of the size of the input stimuli and the range of the lateral interaction function. Additionally, we show that the learning rule can be derived via the optimization of a global cost function in a framework of probabilistic output neurons which represent a set of input stimuli by a sparse code.

Algorithms↗

PubMed2000

The ketogenic diet inhibits epileptogenesis in EL mice: a genetic model for idiopathic epilepsy.

PURPOSE: The ketogenic diet (KD) is a high-fat, low-carbohydrate and -protein diet that has been used to treat refractory seizures in children for more than 75 years. However, little is known about how the KD inhibits seizures or its effects on epileptogenesis. Several animal models of epilepsy have responded favorably to KD treatment, but the KD has not been studied in animals with a genetic predisposition to seizures. Here we studied the antiepileptogenic effect of the KD in EL mice, an animal model for human idiopathic epilepsy. METHODS: Young male EL mice (postnatal day 30) were randomly separated into two groups fed ad libitum with either the KD (treated, n = 21) or Agway chow (control, n = 19). The mice were weighed and tested for seizures once per week for a total of 10 weeks. The effects of the KD on plasma levels of ketone bodies and glucose were analyzed at several time points throughout the study. Associative learning was compared between treated and control animals using a water maze. RESULTS: KD treatment delayed seizure onset in young male EL mice by 1 month; however, seizure protection was transient, inasmuch as the treated and control mice experienced a similar number and intensity of seizures after 6 weeks on the diet. Plasma glucose levels and associative learning were similar in the treated and control groups, but the plasma beta-hydroxybutyrate levels were significantly higher in mice on the KD. The level of ketosis, however, was not predictive of seizure protection in EL mice. CONCLUSION: The KD delayed seizure onset in EL mice, suggesting a transient protection against epileptogenesis. The KD did not influence plasma glucose levels or associative learning. Therefore, the EL mouse may serve as a good model to study the antiepileptogenic mechanisms of the KD.

3-Hydroxybutyric Acid↗

PubMed1991

Chance and necessity in ingestive behavior.

This paper begins with the paradox of invariants arising out of random events, and its title is based on the resolution of that paradox in Monod's book Chance and Necessity. In ingestive behavior this paradox arises particularly in the emergence of constancies in intakes from small bouts of behavior. I point out that one resolution of the paradox can be obtained by the use of averaging, eliminating the necessity to search for special detectors of variables whose values appear to be set by small modifications in a variety of control systems. The apparent regulation of body weight or food intake which remains relatively constant in the aggregate while components which enter into them fluctuate widely, is an illustration of the paradox and one possible resolution. I also describe four basic paradigms that could be taken as the main starting points for a comprehensive theory of ingestive behavior. They could be called regulation, learning, reward, and neural control. To some extent each is involved, and it is impossible at present to assign primacy. I conclude by suggesting that the basic tenet of ingestive behavior is that it occurs when there is an excess of excitation over inhibition in the neural systems that control it, and that the various factors which enter into such a calculation can best be represented by a computer model that incorporates all known controls that will make more accurate predictions than can be made with verbal theorizing. Such predictions can then be used to correct or confirm the model as data is collected. Finally, I provide some specific suggestions for activities of the Society for the Study of Ingestive Behavior.

Animals↗

PubMed2003

Women and acquired preparedness: personality and learning implications for alcohol use.

OBJECTIVE: The purpose of this study was to begin the process of examining the validity of the acquired preparedness model of alcohol use for women. The model holds that trait disinhibition influences the formation of alcohol expectancies, which then influence drinking levels. METHOD: College women (N = 290) completed measures of trait disinhibition, positive and negative expectancies for alcohol and drinking measures. RESULTS: Using structural equation modeling, support was found for the hypothesized processes. Cross-sectional analyses were consistent with two hypothesized mediational pathways: disinhibition was associated with increased positive alcohol expectancies and decreased negative alcohol expectancies; both higher positive and lower negative alcohol expectancies correlated with drinking; and disinhibition's association with drinking was significantly reduced when each type of expectancy was added to a prediction model. CONCLUSIONS: Cross-sectional support for this causal model indicates the value of testing it further with longitudinal trials.

Adolescent↗

PubMed2000

The effects of predator learning, forgetting, and recognition errors on the evolution of warning coloration.

This paper demonstrates that the specifics of predator avoidance learning, information loss, and recognition errors may heavily influence the evolution of aposematism. I establish a mathematical model of the change in frequency over time of bright individuals of a distasteful prey species. Warning color spreads through green beard selection as reformulated by Guilford (1990); bright colored forms gain an advantage due to their phenotypic resemblance to other bright forms, which have been sampled by the predator. I use a general classical conditioning model to examine gradual predator learning and forgetting, and then consider the extreme of one-trial learning and no forgetting over time that may occur with very toxic prey. The advantage of conspicuous coloration under these latter conditions depends upon its role in lowering a constant probability of the prey being misidentified and thus mistakenly attacked by a predator, a rarely emphasized factor in the evolution of warning coloration. This constant probability of mistaken attacks can also be interpreted as a constant probability that forgetting has occurred (forgetting does not increase with time) or a periodic decision by the predator to resample avoided prey. I show that when predators learn and forget gradually, as under the general classical conditioning model, it is very difficult for aposematic coloration to become established unless bright individuals cross an often high threshold frequency through chance factors. In contrast, the conditions expected with highly toxic prey promote the evolution of warning coloration more easily, by means from the fixation of very bright mutations to the fixation of successive mutations each of which causes a small increase in a prey's conspicuousness. The results therefore predict that aposematic coloration may have evolved in a different manner in different predator and prey systems. They also suggest that it may be extremely difficult for warning coloration to evolve in more mildly toxic or distasteful prey outside of a mimicry system.

Animals↗

PubMed2005

Structure and strength in causal induction.

We present a framework for the rational analysis of elemental causal induction-learning about the existence of a relationship between a single cause and effect-based upon causal graphical models. This framework makes precise the distinction between causal structure and causal strength: the difference between asking whether a causal relationship exists and asking how strong that causal relationship might be. We show that two leading rational models of elemental causal induction, DeltaP and causal power, both estimate causal strength, and we introduce a new rational model, causal support, that assesses causal structure. Causal support predicts several key phenomena of causal induction that cannot be accounted for by other rational models, which we explore through a series of experiments. These phenomena include the complex interaction between DeltaP and the base-rate probability of the effect in the absence of the cause, sample size effects, inferences from incomplete contingency tables, and causal learning from rates. Causal support also provides a better account of a number of existing datasets than either DeltaP or causal power.

Adult↗

PubMed2025

Multimodal deep learning for immunotherapy response prediction and biomarker discovery in non-small cell lung cancer.

OBJECTIVE: Immunotherapy has emerged as a promising treatment for advanced non-small cell lung cancer (NSCLC), but accurately predicting which patients will benefit from it remains a major clinical challenge. To address this, we aim to develop a novel multimodal method, DeepAFM, that integrates histopathology, genomic features, and clinical information to predict patient responses to anti-PD-(L)1 immunotherapy. MATERIALS AND METHODS: A total of 93 patients with advanced NSCLC were included in this study. Histopathological whole-slide images were processed using a self-supervised VQVAE2 for representation learning. PCA and K-means clustering were then applied for dimensionality reduction and feature grouping. Key regions of interest were visualized through permutation importance evaluation and color-coding techniques. The extracted histopathological features, along with genomic alterations and clinical variables, were integrated into the DeepAFM multimodal prediction model. RESULTS: The DeepAFM achieved a high predictive performance with an area under the curve (AUC) of 0.77 (95% confidence interval: 0.69-1.00). Attention-based heatmaps revealed that the model could identify critical pathological patterns, genomic mutations, and clinical indicators associated with patient responses to immunotherapy. DISCUSSION: The integration of multimodal data enabled the model to capture complex interactions among pathology, genomics, and clinical characteristics, enhancing the interpretability and predictive power of immunotherapy response prediction. The visualization techniques facilitated the identification of biologically meaningful features and potential biomarkers. CONCLUSION: This study demonstrates the effectiveness of the DeepAFM in predicting responses to immunotherapy in advanced NSCLC. The approach not only improves prediction accuracy but also provides valuable insights for personalized treatment strategies and biomarker discovery.

Humans↗

PubMed2000

Motivational Beliefs, Study Strategies, and Mathematics Attainment in High- and Low-Achieving Chinese Secondary School Students.

In order to examine the relationship between cognitive and motivational variables and their relationship to mathematics attainment, Hong Kong-Chinese students enrolled in schools for high-, average-, and low-achievers completed questionnaires in Year 10 and in Year 11. Low-achievers perceived academic learning as being less useful over time and reported spending less time studying in Year 10 than in Year 11 but high- and low-achievers did not differ on their use of self-regulated learning strategies. Performance on the public examination in mathematics was predicted by prior achievement and Self-Concept of Mathematics Ability. Results underscore the importance of considering cultural beliefs systems and educational systems in models of academic motivation. Copyright 2000 Academic Press.

Journal Article↗