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 1,225 records · Page 68Linked to original sources

PubMed1995

The learning curve for laparoscopic cholecystectomy. The Southern Surgeons Club.

BACKGROUND: The use of laparoscopic surgical procedures without previous training has grown rapidly. At the same time, there have been allegations of increased complications among less experienced surgeons. METHODS: Using multivariate regression analyses, we evaluated the relationship between bile duct injury rate and experience with laparoscopic cholecystectomy for surgeons in the Southern Surgeons Club. RESULTS: Fifty-five surgeons performed 8,839 procedures. Fifteen bile duct injuries (by 13 surgeons) resulted with 90% of the injuries occurring within the first 30 cases performed by an individual surgeon. Multivariate analyses indicated that the only significant factor associated with an adverse outcome was the surgeon's experience with the procedure. A regression model predicted that a surgeon had a 1.7% chance of a bile duct injury occurring in the first case and a 0.17% chance of a bile duct injury at the 50th case. CONCLUSIONS: While surgeons appear to learn this procedure rapidly, institutions might consider requiring surgeons to move beyond the initial learning curve before awarding privileges.

Bile Ducts↗

PubMed2026

NeuroOmics-Net: An interpretable multimodal deep learning framework for Alzheimer's disease diagnosis and progression prediction using neuroimaging, EEG, and genomic data.

Accurate diagnosis and progression prediction of Alzheimer's disease (AD) remain challenging due to the heterogeneous nature of the disease, which involves structural brain degeneration, electrophysiological dysfunction, and molecular dysregulation. Most existing deep learning approaches rely on a single modality or limited multimodal combinations, thereby failing to capture the complex cross-domain interactions underlying AD progression. Furthermore, the scarcity of large-scale datasets containing synchronized neuroimaging, electrophysiological, and genomic measurements restricts the development of comprehensive multimodal diagnostic systems. To address these challenges, this study proposes NeuroOmics-Net, a multimodal deep learning framework for Alzheimer's disease analysis that integrates structural magnetic resonance imaging (sMRI), electroencephalography (EEG), and gene expression data. The proposed framework combines a Hierarchical Multi-View Encoder (HME) for modality-specific feature extraction, a Cross-Omics Attention Fusion (CAF) module for adaptive integration of complementary biomarkers, and a Disease Progression Graph Learning (DPGL) module for modeling progression-related relationships across biological domains. To facilitate cross-modal integration from independent cohorts, Regularized Canonical Correlation Analysis (RCCA) is employed to align heterogeneous feature representations within a shared latent space. Experiments were conducted using publicly available datasets from ADNI, PhysioNet, and GEO repositories comprising 1120 diagnosis-aligned samples. The proposed framework achieved 94.3% classification accuracy and an AUC of 0.975 for distinguishing normal controls (NC), mild cognitive impairment (MCI), and Alzheimer's disease subjects, while attaining 93.7% accuracy for predicting conversion from stable mild cognitive impairment (sMCI) to progressive mild cognitive impairment (pMCI). However, a fairness sensitivity analysis using stratified demographic reweighting revealed accuracy ranging from 90.8% (low-education, high-comorbidity proxy subgroup) to 96.1% (low-risk, high-reserve proxy subgroup), a demographic parity gap of 5.3 percentage points, indicating that overall accuracy reflects a performance ceiling in a relatively homogeneous research cohort rather than a realistic estimate for demographically diverse clinical populations. Comparative evaluations demonstrated consistent improvements over state-of-the-art unimodal and multimodal deep learning models. Interpretability analysis further identified clinically relevant biomarkers, including hippocampal and entorhinal atrophy, theta-alpha EEG alterations, and APOE-associated molecular pathways. Because sMRI, EEG, and gene expression data were sourced from separate, unpaired cohorts with no subjects possessing all three synchronized measurements, all reported cross-modal associations reflect population-level statistical correspondence across diagnosis-matched groups rather than within-subject physiological coupling; no claim of intra-individual causal cross-modal interaction is made. These findings demonstrate that NeuroOmics-Net provides an effective computer-aided framework for multimodal biomedical data processing and Alzheimer's disease analysis. By integrating neuroimaging, electrophysiological, and genomic information, the proposed approach enables accurate diagnosis, progression prediction, and biologically interpretable decision support for clinical and translational applications.

Humans↗

PubMed2006

Predictive data mining in clinical medicine: current issues and guidelines.

BACKGROUND: The widespread availability of new computational methods and tools for data analysis and predictive modeling requires medical informatics researchers and practitioners to systematically select the most appropriate strategy to cope with clinical prediction problems. In particular, the collection of methods known as 'data mining' offers methodological and technical solutions to deal with the analysis of medical data and construction of prediction models. A large variety of these methods requires general and simple guidelines that may help practitioners in the appropriate selection of data mining tools, construction and validation of predictive models, along with the dissemination of predictive models within clinical environments. PURPOSE: The goal of this review is to discuss the extent and role of the research area of predictive data mining and to propose a framework to cope with the problems of constructing, assessing and exploiting data mining models in clinical medicine. METHODS: We review the recent relevant work published in the area of predictive data mining in clinical medicine, highlighting critical issues and summarizing the approaches in a set of learned lessons. RESULTS: The paper provides a comprehensive review of the state of the art of predictive data mining in clinical medicine and gives guidelines to carry out data mining studies in this field. CONCLUSIONS: Predictive data mining is becoming an essential instrument for researchers and clinical practitioners in medicine. Understanding the main issues underlying these methods and the application of agreed and standardized procedures is mandatory for their deployment and the dissemination of results. Thanks to the integration of molecular and clinical data taking place within genomic medicine, the area has recently not only gained a fresh impulse but also a new set of complex problems it needs to address.

Clinical Medicine↗

PubMed2005

A learning rule for place fields in a cortical model: theta phase precession as a network effect.

We show that a model of the hippocampus introduced recently by Scarpetta et al. (2002, Neural Computation 14(10):2371-2396) explains the theta phase precession phenomena. In our model, the theta phase precession comes out as a consequence of the associative-memory-like network dynamics, i.e., the network's ability to imprint and recall oscillatory patterns, coded both by phases and amplitudes of oscillation. The learning rule used to imprint the oscillatory states is a natural generalization of that used for static patterns in the Hopfield model, and is based on the spike-time-dependent synaptic plasticity, experimentally observed.In agreement with experimental findings, the place cells' activity appears at consistently earlier phases of subsequent cycles of the ongoing theta rhythm during a pass through the place field, while the oscillation amplitude of the place cells' firing rate increases as the animal approaches the center of the place field and decreases as the animal leaves the center. The total phase precession of the place cell is lower than 360 degrees , in agreement with experiments. As the animal enters a receptive field, the place cells' activity comes slightly less than 180 degrees after the phase of maximal pyramidal cell population activity, in agreement with the findings of Skaggs et al. (1996, Hippocampus 6:149-172). Our model predicts that the theta phase is much better correlated with location than with time spent in the receptive field. Finally, in agreement with the recent experimental findings of Zugaro et al. (2005, Nature Neuroscience 9(1):67-71), our model predicts that theta phase precession persists after transient intrahippocampal perturbation.

Action Potentials↗

PubMed2005

Epistemological beliefs and approaches to learning: their change through secondary school and their influence on academic performance.

BACKGROUND: In recent decades, two lines of research, phenomenographic and meta-cognitive, have examined students' approaches and epistemological beliefs about learning. To date there has been very little research describing the change in epistemological beliefs in European secondary students, or analysing interrelationships between epistemological beliefs and approaches in order to explain their influence on academic performance. AIMS: The first aim of this investigation is to analyse the change in epistemological beliefs and learning approaches in secondary students as they progress through their studies. The second aim is to examine the effects of epistemological beliefs on learning approaches, and learning approaches on academic performance. SAMPLE: About 1,600 Spanish students, boys and girls, from several secondary schools took part in the study. They were between 12 and 20 years old and their average age was 14.79 years. METHODS: Measures of epistemological beliefs (EQ: Epistemological Questionnaire), learning approaches (LPQ: Learning Process Questionnaire), and academic performance were obtained. Confirmatory factor analysis was used to examine the dimensionality of the EQ and LPQ questionnaires. In order to achieve our two aims, different statistical techniques were used: MANOVA and ANOVA for our first aim, and structural equation modelling for our second aim. RESULTS: Throughout secondary education epistemological beliefs undergo change, becoming more realistic and complex, and deep-approach scores decline significantly. It was shown that, as predicted, epistemological beliefs influenced academic achievement directly, and also indirectly via students' learning approaches. CONCLUSIONS: Our findings point to two conclusions. First, epistemological beliefs and learning approaches change as pupils advance in their studies. Second, the relationship between epistemological beliefs and academic achievement is mediated by approaches to learning.

Achievement↗

PubMed2004

Neuroscience. Addiction as compulsive reward prediction.

Explore the source record for details and available documents.

Animals↗

PubMed2000

The impact of deep brain stimulation on executive function in Parkinson's disease.

Deep brain stimulation (DBS) of the subthalamic nucleus (STN) or the internal segment of the globus pallidus (GPi) improves Parkinson's disease and increases frontal blood flow. We assessed the effects of bilateral DBS on executive function in Parkinson's disease patients, seven with electrodes implanted in the STN and six in the GPi. Patients were assessed off medication with stimulators off, on and off again. The groups showed differential change with stimulation on the Reitan Trail-Making test (TMT B) (STN more improved) and on some measures of random number generation and Wisconsin Card Sorting (STN improved, GPi worse with stimulation). Across the groups, stimulation speeded up responding (Stroop control trial, TMT A) and improved performance on paced serial addition and missing digit tests. Conversely, conditional associative learning became more errorful with stimulation across the two groups. In general, change in performance with stimulation was significant for the STN but not the GPi group. These results support two opposite predictions. In support of current models of Parkinson's disease, 'releasing the brake' on frontal function with DBS improved aspects of executive function. Conversely, disruption of basal ganglia outflow during DBS impaired performance on tests requiring changing behaviour in novel contexts as predicted by Marsden and Obeso in 1994.

Association Learning↗

PubMed2006

TICO: a tool for postprocessing the predictions of prokaryotic translation initiation sites.

Exact localization of the translation initiation sites (TIS) in prokaryotic genomes is difficult to achieve using conventional gene finders. We recently introduced the program TICO for postprocessing TIS predictions based on a completely unsupervised learning algorithm. The program can be utilized through our web interface at http://tico.gobics.de/ and it is also freely available as a commandline version for Linux and Windows. The latest version of our program provides a tool for visualization of the resulting TIS model. Although the underlying method is not based on any specific assumptions about characteristic sequence features of prokaryotic TIS the prediction rates of our tool are competitive on experimentally verified test data.

Algorithms↗

PubMed1997

Prediction of complex two-dimensional trajectories by a cerebellar model of smooth pursuit eye movement.

A neural network model based on the anatomy and physiology of the cerebellum is presented that can generate both simple and complex predictive pursuit, while also responding in a feedback mode to visual perturbations from an ongoing trajectory. The model allows the prediction of complex movements by adding two features that are not present in other pursuit models: an array of inputs distributed over a range of physiologically justified delays, and a novel, biologically plausible learning rule that generated changes in synaptic strengths in response to retinal slip errors that arrive after long delays. To directly test the model, its output was compared with the behavior of monkeys tracking the same trajectories. There was a close correspondence between model and monkey performance. Complex target trajectories were created by summing two or three sinusoidal components of different frequencies along horizontal and/or vertical axes. Both the model and the monkeys were able to track these complex sum-of-sines trajectories with small phase delays that averaged 8 and 20 ms in magnitude, respectively. Both the model and the monkeys showed a consistent relationship between the high- and low-frequency components of pursuit: high-frequency components were tracked with small phase lags, whereas low-frequency components were tracked with phase leads. The model was also trained to track targets moving along a circular trajectory with infrequent right-angle perturbations that moved the target along a circle meridian. Before the perturbation, the model tracked the target with very small phase differences that averaged 5 ms. After the perturbation, the model overshot the target while continuing along the expected nonperturbed circular trajectory for 80 ms, before it moved toward the new perturbed trajectory. Monkeys showed similar behaviors with an average phase difference of 3 ms during circular pursuit, followed by a perturbation response after 90 ms. In both cases, the delays required to process visual information were much longer than delays associated with nonperturbed circular and sum-of-sines pursuit. This suggests that both the model and the eye make short-term predictions about future events to compensate for visual feedback delays in receiving information about the direction of a target moving along a changing trajectory. In addition, both the eye and the model can adjust to abrupt changes in target direction on the basis of visual feedback, but do so after significant processing delays.

Animals↗

PubMed1984

Random-walk and accumulator models of psychophysical discrimination: a critical evaluation.

The accumulator model proposed by Vickers and the modified random-walk model proposed by Link and Heath are compared in their ability to account for confidence judgments in line-length discrimination tasks. The random-walk model proves to be a viable alternative to the accumulator model, and is able to account for the relationship between mean response time and confidence. The parameter estimation techniques available for the random-walk model are considered advantageous when compared with the accumulator model, because the predictions from the latter have been obtained with the use of computer simulation.

Computers↗

PubMed2000

Assessing the accuracy of prediction algorithms for classification: an overview.

We provide a unified overview of methods that currently are widely used to assess the accuracy of prediction algorithms, from raw percentages, quadratic error measures and other distances, and correlation coefficients, and to information theoretic measures such as relative entropy and mutual information. We briefly discuss the advantages and disadvantages of each approach. For classification tasks, we derive new learning algorithms for the design of prediction systems by directly optimising the correlation coefficient. We observe and prove several results relating sensitivity and specificity of optimal systems. While the principles are general, we illustrate the applicability on specific problems such as protein secondary structure and signal peptide prediction.

Algorithms↗

PubMed2004

Pairs do not suffer interference from other types of pairs or single items in associative recognition.

What is the source of interference on a memory test following study of a list containing different types of pairs? Many current models predict that pairs and singles of all types will jointly interfere and therefore harm memory. Such list length effects have often been observed for lists of a single-item type (e.g., a list of words). Here, we examine interference for lists containing multiple types of pairs (e.g., word-word, face-face, word-face). In three experiments, we manipulate the number of each type on the study list. In associative recognition, discrimination fell as the number of pairs of the same type rose, but the number of pairs of other types had little effect. That is, we found a list length effect within, but not between, classes of stimuli. We highlight the importance of representation and propose alternatives to current model representations that can predict such findings.

Discrimination, Psychological↗

PubMed2004

A hidden markov model derived structural alphabet for proteins.

Understanding and predicting protein structures depends on the complexity and the accuracy of the models used to represent them. We have set up a hidden Markov model that discretizes protein backbone conformation as series of overlapping fragments (states) of four residues length. This approach learns simultaneously the geometry of the states and their connections. We obtain, using a statistical criterion, an optimal systematic decomposition of the conformational variability of the protein peptidic chain in 27 states with strong connection logic. This result is stable over different protein sets. Our model fits well the previous knowledge related to protein architecture organisation and seems able to grab some subtle details of protein organisation, such as helix sub-level organisation schemes. Taking into account the dependence between the states results in a description of local protein structure of low complexity. On an average, the model makes use of only 8.3 states among 27 to describe each position of a protein structure. Although we use short fragments, the learning process on entire protein conformations captures the logic of the assembly on a larger scale. Using such a model, the structure of proteins can be reconstructed with an average accuracy close to 1.1A root-mean-square deviation and for a complexity of only 3. Finally, we also observe that sequence specificity increases with the number of states of the structural alphabet. Such models can constitute a very relevant approach to the analysis of protein architecture in particular for protein structure prediction.

Algorithms↗

PubMed2005

[New trends in the evaluation of mathematics learning disabilities. The role of metacognition].

INTRODUCTION: The current trends in the evaluation of mathematics learning disabilities (MLD), based on cognitive and empirical models, are oriented towards combining procedures involving the criteria and the evaluation of cognitive and metacognitive processes, associated to performance in mathematical tasks. AIMS: The objective of this study is to analyse the metacognitive skills of prediction and evaluation in performing maths tasks and to compare metacognitive performance among pupils with MLD and younger pupils without MLD, who have the same level of mathematical performance. Likewise, we analyse these pupils' desire to learn. Subjects and methods. We compare a total of 44 pupils from the second cycle of primary education (8-10 years old) with and without mathematics learning disabilities. RESULTS: Significant differences are observed between pupils with and without mathematics learning disabilities in their capacity to predict and assess all of the tasks evaluated. As regards their 'desire to learn', no significant differences were found between pupils with and without MLD, which indicated that those with MLD assess their chances of successfully performing maths tasks in the same way as those without MLD. Finally, the findings reveal a similar metacognitive profile in pupils with MLD and the younger pupils with no mathematics learning disabilities. CONCLUSIONS: In future studies we consider it important to analyse the influence of the socio-affective belief system in the use of metacognitive skills.

Learning Disabilities↗

PubMed1995

A proposed structural model of domain 1 of fasciclin III neural cell adhesion protein based on an inverse folding algorithm.

Fasciclin III is an integral membrane protein expressed on a subset of axons in the developing Drosophila nervous system. It consists of an intracellular domain, a transmembrane region, and an extracellular region composed of three domains, each predicted to form an immunoglobulin-like fold. The most N-terminal of these domains is expected to be important in mediating cell-cell recognition events during nervous system development. To learn more about the structure/function relationships in this cellular recognition molecule, a model structure of this domain was built. A sequence-to-structure alignment algorithm was used to align the protein sequence of the fasciclin III first domain to the immunoglobulin McPC603 structure. Based on this alignment, a model of the domain was built using standard homology modeling techniques. Side-chain conformations were automatically modeled using a rotamer search algorithm and the model was minimized to relax atomic overlaps. The resulting model is compact and has chemical characteristics consistent with related globular protein structures. This model is a de novo test of the sequence-to-structure alignment algorithm and is currently being used as the basis for mutagenesis experiments to discern the parts of the fasciclin III protein that are necessary for homophilic molecular recognition in the developing Drosophila nervous system.

Algorithms↗

PubMed2026

Serum Proteomic Profiling Implicates a Dysregulated Neurohormonal-Inflammatory Axis in Post-Fontan Sinus Tachycardia.

BACKGROUND: Postoperative sinus tachycardia is a poorly understood complication following the Fontan procedure. The molecular signaling cascades triggering acute tachycardia remain uncharacterized, limiting therapeutic innovation. Here, we present a retrospective study leveraging serum proteomics and machine learning to identify the molecular drivers of postoperative Fontan sinus tachycardia. METHODS: We integrated a clinically relevant ovine Fontan model with continuous telemetric heart rate monitoring and human patient data. Serum proteomics coupled with least absolute shrinkage and selection operator and Boruta machine learning algorithms were used to identify protein panels predictive of postoperative sinus tachycardia. Cross-species validation was performed by comparing proteomic signatures from sheep and pediatric patients undergoing Glenn or Fontan surgery. RESULTS: Ovine Fontan animals demonstrated significant heart rate elevation beginning on postoperative day 1, peaking at postoperative day 3 (159.4±11.7 bpm versus preoperative, 105.3±10.5 bpm; P=0.0002), before trending toward baseline by postoperative day 10. This pattern was mirrored in human patients with a more modest magnitude. Surgical controls did not exhibit tachycardia. The principal component most correlated with heart rate (principal component 1: r=0.78, P=2.2×10-4) was enriched for inflammatory and neural pathways. The Boruta algorithm identified an 11-protein panel with strong predictive power (area under the receiver operating characteristic curve, 0.963). Cross-species comparison demonstrated that angiotensinogen, angiotensin-converting enzyme, and pentraxin 3 were similarly dysregulated in both species postoperatively. CONCLUSIONS: This study provides molecular evidence implicating a dysregulated neurohormonal-inflammatory axis in acute postoperative Fontan sinus tachycardia and establishes a foundation for developing targeted diagnostics and therapeutics for this complication.

Animals↗

PubMed2004

Evidence accumulation in decision making: unifying the "take the best" and the "rational" models.

An evidence accumulation model of forced-choice decision making is proposed to unify the fast and frugal take the best (TTB) model and the alternative rational (RAT) model with which it is usually contrasted. The basic idea is to treat the TTB model as a sequential-sampling process that terminates as soon as any evidence in favor of a decision is found and the rational approach as a sequential-sampling process that terminates only when all available information has been assessed. The unified TTB and RAT models were tested in an experiment in which participants learned to make correct judgments for a set of real-world stimuli on the basis of feedback, and were then asked to make additional judgments without feedback for cases in which the TTB and the rational models made different predictions. The results show that, in both experiments, there was strong intraparticipant consistency in the use of either the TTB or the rational model but large interparticipant differences in which model was used. The unified model is shown to be able to capture the differences in decision making across participants in an interpretable way and is preferred by the minimum description length model selection criterion.

Adult↗

PubMedDate not supplied

ADME evaluation in drug discovery. 4. Prediction of aqueous solubility based on atom contribution approach.

A novel method for the estimation of aqueous solubility was solely based on simple atom contribution. Each atom in a molecule has its own contribution to aqueous solubility and was developed. Altogether 76 atom types were used to classify atoms with different chemical environments. Moreover, two correction factors, including hydrophobic carbon and square of molecular weight, were used to account for the inter-/intramolecular hydrophobic interactions and bulkiness effect. The contribution coefficients of different atom types and correction factors were generated based on a multiple linear regression using a learning set consisting of 1290 organic compounds. The obtained linear regression model possesses good statistical significance with an overall correlation coefficient (r) of 0.96, a standard deviation (s) of 0.61, and an unsigned mean error (UME) of 0.48. The actual prediction potential of the model was validated through an external test set with 21 pharmaceutically and environmentally interesting compounds. For the test set, a predictive r=0.94, s=0.84, and UME=0.52 were achieved. Comparisons among eight procedures of solubility calculation for those 21 molecules demonstrate that our model bears very good accuracy and is comparable to or even better than most reported techniques based on molecular descriptors. Moreover, we compared the performance of our model to a test set of 120 molecules with a popular group contribution method developed by Klopman et al. For this test set, our model gives a very effective prediction (r=0.96, s=0.79, UME=0.57), which is obviously superior to the predicted results (r=0.96, s=0.84, UME=0.70) given by the Klopman's group contribution approach. Because of the adoption of atoms as the basic units, our addition model does not contain a "missing fragment" problem and thus may be more simple and universal than the group contribution models and can give predictions for any organic molecules. A program, drug-LOGS, had been developed to identify the occurrence of atom types and estimate the aqueous solubility of a molecule.

Drug Design↗