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 1,117 records · Page 62Linked to original sources

PubMed1992

Neural networks in pharmacodynamic modeling. Is current modeling practice of complex kinetic systems at a dead end?

Neural networks (NN) are computational systems implemented in software or hardware that attempt to simulate the neurological processing abilities of biological systems, in particular the brain. Computational NN are classified as parallel distributed processing systems that for many tasks are recognized to have superior processing capability to the classical sequential Von Neuman computer model. NN are recognized mainly in terms of their adaptive learning and self-organization features and their nonlinear processing capability and are considered most suitable to deal with complex multivariate systems that are poorly understood and difficult to model by classical inductive, logically structured modeling techniques. A NN is applied to demonstrate one of the potentially many applications of NN for modeling complex kinetic systems. The NN was used to predict the effect of alfentanil on the heart rate resulting from a complex infusion scheme applied to six rabbits. Drug input-drug effect data resulting from a repeated, triple infusion rate scheme lasting from 30 to 180 min was used to train the NN to recognize and emulate the input-effect behavior of the system. With the NN memory fixed from the 30- to 180-min learning phase the NN was then tested for its ability to predict the effect resulting from a multiple infusion rate scheme applied in the subsequent 180 to 300 min of the experiment. The NN's ability to emulate the system (30-180 min) was excellent and its predictive extrapolation capability (180-300 min) was very good (mean relative prediction accuracy of 78%). The NN was best in predicting the higher intensity effect and was able to identify and predict an overshoot phenomenon likely caused by a withdrawal effect from acute tolerance. Current modeling philosophy and practice is discussed on the basis of the alternative offered by NN in the modeling of complex kinetic systems. In modeling such systems it is questioned whether traditional modeling practice that insists on structure relevance and conceptually pleasing structures has any practical advantages over the empirical NN approach that largely ignores structure relevance but concentrates on the emulation of the behavior of the kinetic system. The traditional searching for appropriate models of complex kinetic systems is a painstakingly slow process. In contrast, the search for empirical models using NN will continue to improve, limited only by technological advances supporting the very promising NN developments.

Alfentanil↗

PubMed2002

N-terminal N-myristoylation of proteins: prediction of substrate proteins from amino acid sequence.

Myristoylation by the myristoyl-CoA:protein N-myristoyltransferase (NMT) is an important lipid anchor modification of eukaryotic and viral proteins. Automated prediction of N-terminal N-myristoylation from the substrate protein sequence alone is necessary for large-scale sequence annotation projects but it requires a low rate of false positive hits in addition to a sufficient sensitivity. Our previous analysis of substrate protein sequence variability, NMT sequences and 3D structures has revealed motif properties in addition to the known PROSITE motif that are utilized in a new predictor described here. The composite prediction function (with separate ad hoc parameterization (a) for queries from non-fungal eukaryotes and their viruses and (b) for sequences from fungal species) consists of terms evaluating amino acid type preferences at sequences positions close to the N terminus as well as terms penalizing deviations from the physical property pattern of amino acid side-chains encoded in multi-residue correlation within the motif sequence. The algorithm has been validated with a self-consistency and two jack-knife tests for the learning set as well as with kinetic data for model substrates. The sensitivity in recognizing documented NMT substrates is above 95 % for both taxon-specific versions. The corresponding rate of false positive prediction (for sequences with an N-terminal glycine residue) is close to 0.5 %; thus, the technique is applicable for large-scale automated sequence database annotation. The predictor is available as public WWW-server with the URL http://mendel.imp.univie.ac.at/myristate/. Additionally, we propose a version of the predictor that identifies a number of proteolytic protein processing sites at internal glycine residues and that evaluates possible N-terminal myristoylation of the protein fragments.A scan of public protein databases revealed new potential NMT targets for which the myristoyl modification may be of critical importance for biological function. Among others, the list includes kinases, phosphatases, proteasomal regulatory subunit 4, kinase interacting proteins KIP1/KIP2, protozoan flagellar proteins, homologues of mitochondrial translocase TOM40, of the neuronal calcium sensor NCS-1 and of the cytochrome c-type heme lyase CCHL. Analyses of complete eukaryote genomes indicate that about 0.5 % of all encoded proteins are apparent NMT substrates except for a higher fraction in Arabidopsis thaliana ( approximately 0.8 %).

Acyltransferases↗

PubMed2005

A model for stochastic drift in memory strength to account for judgments of learning.

Previous research has shown that judgments of learning (JOLs) made immediately after encoding have a low correlation with actual cued-recall performance, whereas the correlation is high for delayed judgments. In this article, the authors propose a formal theory describing the stochastic drift of memory strength over the retention interval to account for the delayed-JOL effect. This is done by first decomposing the aggregated memory strength into exponential functions with slow and fast memory traces. The mean aggregated memory strength shows power-function forgetting curves. The drift of the memory strength is large for immediate JOLs (causing a low predictability) and weak for delayed JOLs (causing a high predictability). Consistent with empirical data, the model makes a novel prediction of JOL asymmetry, or that immediate weak JOLs are more predictive of future performance than are immediate strong JOLs. The JOL distributions for immediate and delayed JOLs are also accounted for.

Cues↗

PubMed2003

The word-frequency paradox for recall/recognition occurs for pictures.

A yes-no recognition task and two recall tasks were conducted using pictures of high and low familiarity ratings. Picture familiarity had analogous effects to word frequency, and replicated the word-frequency paradox in recall and recognition. Low-familiarity pictures were more recognizable than high-familiarity pictures, pure lists of high-familiarity pictures were more recallable than pure lists of low-familiarity pictures, and there was no effect of familiarity for mixed lists. These results are consistent with the predictions of the Search of Associative Memory (SAM) model.

Adolescent↗

PubMed1998

Multiple paired forward and inverse models for motor control.

Humans demonstrate a remarkable ability to generate accurate and appropriate motor behavior under many different and often uncertain environmental conditions. In this paper, we propose a modular approach to such motor learning and control. We review the behavioral evidence and benefits of modularity, and propose a new architecture based on multiple pairs of inverse (controller) and forward (predictor) models. Within each pair, the inverse and forward models are tightly coupled both during their acquisition, through motor learning, and use, during which the forward models determine the contribution of each inverse model's output to the final motor command. This architecture can simultaneously learn the multiple inverse models necessary for control as well as how to select the inverse models appropriate for a given environment. Finally, we describe specific predictions of the model, which can be tested experimentally.

Journal Article↗

PubMed2001

Processing and representation of meta-data for sleep apnea diagnosis with an artificial intelligence approach.

In this article, we revise and try to resolve some of the problems inherent in questionnaire screening of sleep apnea cases and apnea diagnosis based on attributes which are relevant and reliable. We present a way of learning information about the relevance of the data, comparing this with the definition of the information by the medical expert. We generate a predictive data model using a data aggregation operator which takes relevance and reliability information about the data into account to produce a diagnosis for each case. We also introduce a grade of membership for each question response which allows the patient to indicate a level of confidence or doubt in their own judgement. The method is tested with data collected from patients in a Sleep Clinic using questionnaires specially designed for the study. Other artificial intelligence predictive modeling algorithms are also tested on the same data and their predictive accuracy compared to that of the aggregation operator.

Adult↗

PubMed2006

Empirically and clinically useful decision making in psychotherapy: differential predictions with treatment response models.

In the delivery of clinical services, outcomes monitoring (i.e., repeated assessments of a patient's response to treatment) can be used to support clinical decision making (i.e., recurrent revisions of outcome expectations on the basis of that response). Outcomes monitoring can be particularly useful in the context of established practice research networks. This article presents a strategy to disaggregate patients into homogeneous subgroups to generate optimal expected treatment response profiles, which can be used to predict and track the progress of patients in different treatment modalities. The study was based on data from 618 diagnostically diverse patients treated with either a cognitive-behavioral treatment protocol (n = 262) or an integrative cognitive-behavioral and interpersonal treatment protocol (n = 356). The validity of expected treatment response models to predict treatment in those 2 protocols for individual patients was evaluated. The ways such a procedure might be used in outpatient centers to learn more about patients, predict treatment response, and improve clinical practice are discussed.

Adult↗

PubMed2001

SAR modeling of unbalanced data sets.

The increased acceptance of SAR approaches to hazard identification has led us to investigate methods to improve the predictive performance of SAR models. In the present study we demonstrate that although on theoretical grounds the ratio of active to inactive chemicals in the learning set should be unity, SAR models can "tolerate" an unbalanced range in ratios from 3:1 (i.e., 75% actives) to 1:2 (i.e., 33% actives) and still perform adequately. On the other hand SAR models derived from learning sets with ratios in excess of 4:1 (80% actives), even when corrected for the initial ratio do not perform satisfactorily.

Proportional Hazards Models↗

PubMed1994

RNA sequence analysis using covariance models.

We describe a general approach to several RNA sequence analysis problems using probabilistic models that flexibly describe the secondary structure and primary sequence consensus of an RNA sequence family. We call these models 'covariance models'. A covariance model of tRNA sequences is an extremely sensitive and discriminative tool for searching for additional tRNAs and tRNA-related sequences in sequence databases. A model can be built automatically from an existing sequence alignment. We also describe an algorithm for learning a model and hence a consensus secondary structure from initially unaligned example sequences and no prior structural information. Models trained on unaligned tRNA examples correctly predict tRNA secondary structure and produce high-quality multiple alignments. The approach may be applied to any family of small RNA sequences.

Algorithms↗

PubMed2004

The impaired coping induced by early deprivation is reversed by chronic fluoxetine treatment in adult fischer rats.

The symptoms of depression include feelings of reduced coping ability and increased helplessness. Early life adversity increases vulnerability to depression. In rats, the quantification of ability to cope with adverse challenge can be achieved using preexposure to an inescapable aversive stimulus and subsequent assessment of escape or avoidance deficits in the same environment. Here we investigated the predictive validity of a model in which, in the Fischer rat strain, postnatal isolation leads in adulthood to a state of increased sensitivity to develop an escape or avoidance deficit. On days 1-14 rat pups were isolated for 4 hours (early deprivation, ED) or for 15 minutes (early handling, EH), or were left completely undisturbed (non-handling, NH). In adulthood, subjects were placed in a shuttle box and half were exposed to brief, mild foot shocks (preexposure, PE) and the other half were non-preexposed (NPE). Half of the PE and NPE subjects were then treated for 21 days with fluoxetine and the other half with vehicle. In males, although there was no overall preexposure effect on avoidance behaviour, ED-PE and ED-NPE and EH-PE and EH-NPE demonstrated an avoidance deficit relative to NH. Fluoxetine attenuated this deficit and most notably in ED-PE. In females, vehicle ED-PE demonstrated an avoidance deficit relative to NH-PE; fluoxetine attenuated this ED effect. These findings provide supportive evidence for the predictive validity of this depression model.

Adaptation, Psychological↗

PubMed1973

Application of Bower's one-element model to paired-associate learning by pigeons.

Bower's (1961) all-or-none model of human paired-associate learning was applied to individual data supplied by three pigeons. When the center one of three keys was illuminated with red light or with three white dots in a vertical array on a black ground, pecking on the left key was reinforced. When the center key was lighted green or with a horizontal array of three white dots on a black ground, pecking on the right key was reinforced. The left and right keys were illuminated with white light. The task was considered to be analogous to learning a paired-associate list of four pairs involving four stimulus items and two response items. The model was evaluated by comparing the following model predictions with values obtained from each animal: trials-to-criterion, standard deviation of trials-to-criterion, standard deviation of errors-to-criterion, mean error runs, mean error runs of lengths one to four, and autocorrelations of errors of lags one to three. Most of the predictions based upon the model were in close agreement with the obtained data.

Journal Article↗

PubMed2004

A response time model for judging order relationship between two symbolic stimuli.

The response time to judge the order relationship between two symbolic stimuli is frequently modeled as the time spent in a (constant-rate) accumulative sampling process until a threshold is reached. We will show that empirical descriptions of observed effects in number comparisons suggest an accrual process that reaches the threshold at an exponential rate. The model accrual equations and stopping conditions have an immediate interpretation in terms of a simple quantitative connectionist network. The encoded stimuli and thresholds are inputs to the network. The former are considered to result from the participant's learning history, and the latter modulate the rearrangement of the network parts; each arrangement models a different task. We have found a good correlation between model predictions and other authors' experimental data, both in number comparisons and in experiments in which the ordering of the symbolic stimuli has been artificially induced. Incorrect answers are discussed, and predictions are compared with data. We will explore differences and similarities with other approaches, such as random walk and the symbolic comparison model. In a limit case, our model becomes identical to the discriminability model.

Humans↗

PubMed1989

Teenagers' beliefs about AIDS education and physicians' perceptions about them.

A survey of 189 Louisiana teenagers and 80 Louisiana family physicians revealed that the teenagers overwhelmingly preferred to learn about acquired immune deficiency syndrome (AIDS) from a physician. This result stimulated an interest in learning whether family physicians shared adolescents' opinions that they are the best teachers for AIDS education and whether family physicians understand adolescents' knowledge and beliefs about AIDS sufficiently well to be effective AIDS educators. Family physicians' responses to a questionnaire based, in part, on the Health Belief Model were compared with teenagers' responses about their knowledge, health beliefs, and preferred format and method of learning about AIDS. Results indicated that family physicians' predictions about teenagers' knowledge and beliefs about AIDS were not always accurate, but, except when physicians underestimated the teenagers' perceived obstacles to AIDS prevention, the data suggested that physicians would be effective in teaching teenagers about AIDS. Family physicians showed good agreement with teens in estimating their desired method and format for learning about AIDS, including their preference for a physician instructor.

Acquired Immunodeficiency Syndrome↗

PubMed2026

Multimodal alignment improves generalizability of genomic biomarker prediction in computational pathology.

Computational pathology models that use digitized histopathology whole-slide images have the potential to become a cost-effective and scalable alternative to molecular assays for the prediction of genomic biomarkers, a key task in precision oncology. However, as new genomic biomarkers are discovered or quantified, large, labeled datasets must be prospectively collected to train new models. To address this challenge, we developed multimodal alignment for biomarker learning and generalization (MARBLE), a multimodal contrastive pretraining strategy that integrates structured biomarker knowledge into representation learning of histopathology images. MARBLE aligns histopathology-derived representations with representations of genomic biomarkers generated by a large language model (LLM) and a protein language model (PLM). This biologically informed alignment enables data-efficient generalization to novel, out-of-distribution biomarkers. Using the MSK-IMPACT cohort of over 40,000 patients across multiple biomarker panel versions, we design experiments grounded in real-world data to demonstrate the value of our proposed approach.

CP: computational biology↗

PubMed2006

Interfering with theories of sleep and memory: sleep, declarative memory, and associative interference.

Mounting behavioral evidence in humans supports the claim that sleep leads to improvements in recently acquired, nondeclarative memories. Examples include motor-sequence learning; visual-discrimination learning; and perceptual learning of a synthetic language. In contrast, there are limited human data supporting a benefit of sleep for declarative (hippocampus-mediated) memory in humans (for review, see). This is particularly surprising given that animal models (e.g.,) and neuroimaging studies (e.g.,) predict that sleep facilitates hippocampus-based memory consolidation. We hypothesized that we could unmask the benefits of sleep by challenging the declarative memory system with competing information (interference). This is the first study to demonstrate that sleep protects declarative memories from subsequent associative interference, and it has important implications for understanding the neurobiology of memory consolidation.

Adolescent↗

PubMed2005

Construct validity of the animal latent inhibition model of selective attention deficits in schizophrenia.

Latent inhibition (LI) is demonstrated when a previously unattended/inconsequential stimulus is less effective in a new learning situation than a novel stimulus. In rats and humans, LI is reduced by dopamine agonists and increased by dopamine antagonists. In addition, LI is attenuated in actively psychotic schizophrenia patients, thus conferring strong predictive validity to the animal LI preparation for schizophrenia. However, the validity of the attentional construct in the LI model of schizophrenia dysfunction depends on confirming two assumptions: that animal and human LI share a common process, and that the process is related to selective attention. Evidence to support both assumptions is presented, followed by a description of a conditioned attention theory that emphasizes the role of initial levels of attention elicited by repeated relevant and irrelevant stimuli, and the differences between these levels in schizophrenia and normal groups.

Animals↗

PubMedDate not supplied

Porcine transfer study: virtual reality simulator training compared with porcine training in endovascular novices.

PURPOSE: To compare the learning of endovascular interventional skills by training on pig models versus virtual reality simulators. METHODS: Twelve endovascular novices participated in a study consisting of a pig laboratory (P-Lab) and a virtual reality laboratory (VR-Lab). Subjects were stratified by experience and randomized into four training groups. Following 1 hr of didactic instruction, all attempted an iliac artery stenosis (IAS) revascularization in both laboratories. Onsite proctors evaluated performances using task-specific checklists and global rating scales, yielding a Total Score. Participants completed two training sessions of 3 hr each, using their group's assigned method (P-Lab x 2, P-Lab + VR-Lab, VR-Lab + P-Lab, or VR-Lab x 2) and were re-evaluated in both laboratories. A panel of two highly experienced interventional radiologists performed assessments from video recordings. ANCOVA analysis of Total Score against years of surgical, interventional radiology (IR) experience and cumulative number of P-Lab or VR-Lab sessions was conducted. Inter-rater reliability (IRR) was determined by comparing proctored scores with the video assessors in only the VR-Lab. RESULTS: VR-Lab sessions improved the VR-Lab Total Score (beta = 3.029, p = 0.0015) and P-Lab Total Score (beta = 1.814, p = 0.0452). P-Lab sessions increased the P-Lab Total Score (beta = 4.074, p < 0.0001) but had no effect on the VR-Lab Total Score. In the general statistical model, both P-Lab sessions (beta = 2.552, p = 0.0010) and VR-Lab sessions (beta = 2.435, p = 0.0032) significantly improved Total Score. Neither previous surgical experience nor IR experience predicted Total Score. VR-Lab scores were consistently higher than the P-Lab scores (Delta = 6.659, p < 0.0001). VR-Lab IRR was substantial (r = 0.649, p < 0.0008). CONCLUSIONS: Endovascular skills learned in the virtual environment may be transferable to the real catheterization laboratory as modeled in the P-Lab.

Adult↗

PubMed1988

A graph-dynamic model of the power law of practice and the problem-solving fan-effect.

Numerous human learning phenomena have been observed and captured by individual laws, but no unified theory of learning has succeeded in accounting for these observations. A theory and model are proposed that account for two of these phenomena: the power law of practice and the problem-solving fan-effect. The power law of practice states that the speed of performance of a task will improve as a power of the number of times that the task is performed. The power law resulting from two sorts of problem-solving changes, addition of operators to the problem-space graph and alterations in the decision procedure used to decide which operator to apply at a particular state, is empirically demonstrated. The model provides an analytic account for both of these sources of the power law. The model also predicts a problem-solving fan-effect, slowdown during practice caused by an increase in the difficulty of making useful decisions between possible paths, which is also found empirically.

Decision Making↗