SEARCH · PubMed Health

Results for “Multimodal learning”

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 235 records · Page 13Linked to original sources

PubMed1994

Predicting outcome in a multimodal day treatment program for children with severe behaviour problems.

The demand for day treatment programs for children with severe behaviour problems has greatly increased in the past decade. The objective of this study is to identify characteristics of children who benefit most from such a program. The sample consisted of 63 children, aged five to 13, assessed at admission and discharge from the program. Outcome measures included behavioural improvement and school reintegration. At admission, multivariate analyses revealed that children who made the greatest behavioural improvements had less severe behaviour problems, lower IQ scores and more disturbed family functioning. Children reintegrated into regular classes after discharge were younger at admission, had better reading skills, no attention deficit hyperactivity disorder and little parental marital discord. Findings suggest that successful outcome can be predicted from characteristics of children at intake. A multimodal approach is needed to address the different problems areas in a child's life.

Adolescent↗

PubMed1982

Photic cuing of escape by rats from an intense microwave field.

A total of 16 female hooded rats was first observed for baseline behaviors and then they received 25 2-min trials of training, five trials per day, under one of four stimulus conditions (all ns = 4): exposure to a highly intense 918-MHz field (dose rate, 60 mW/g); exposure to photic stimulation (approximately 350 1x); exposure to the field in synchrony with photic stimulation; or exposure to faradic shock (approximately 800 micro A rms). During conditioning trials, which were separated by 2-min intertrial intervals, entry by a rat into a safe area of a multimode cavity resulted in immediate and complete cessation of stimulation; exit, in resumption. Acquisition of the escape response was rapid and highly efficient for shocked animals and was less rapid and efficient but was reliably demonstrated by irradiated animals that were also signaled by light. In the absence of microwave irradiation, cessation of light did not reliably motivate escape behavior. Although there was weak evidence of escape learning by rats subjected only to microwave irradiation, their performance failed to differ reliably from those of rats in the light-only condition. These data confirm and extend those of Carroll et al, which indicate that potentially lethal, deeply penetrating, nonpulsed microwaves in a multipath field lack the sensory quality to motivate efficient aversive behavior by the rat.

Animals↗

PubMed2026

HXMS: a standardized file format for HX-MS data.

MOTIVATION: Hydrogen/deuterium exchange-mass spectrometry (HX-MS) is a rapidly expanding technique used to investigate protein conformational ensembles. The growing popularity and utility of HX-MS has driven the development of diverse instrumentation and software, resulting in inconsistent, non-standardized data analysis and representation. Most HX-MS data formats also employ only mean deuteration representations of the data rather than full isotopic mass spectra, which reduces the information content of the data and limits downstream quantitative analysis. RESULTS: Inspired by reliable protein structure and genomics data formats, we present HXMS, a unified, lightweight, scalable, and human-readable file format for HX-MS data. The HXMS format preserves the isotopic mass envelopes for all peptides, captures the full experimental time-course including fully deuterated control samples, and contains all other key information. It supports multimodal distributions, post-translational modifications (PTMs), and experimental replicates. To promote compatibility with existing HX-MS workflows, we also developed PFLink, a Python package that converts exported data files from commonly used HX-MS software to the HXMS format. PFLink and the HXMS format will enable quantitative, higher-resolution data processing, improved data sharing and storage among HX-MS practitioners, future machine learning applications, and further developments in HX-MS analysis. AVAILABILITY AND IMPLEMENTATION: PFLink is publicly available to install locally on HuggingFace, alongside documentation, or use online at HuggingFace (https://huggingface.co/spaces/glasgow-lab/PFlink). The supplementary information includes sample input files, sample HXMS files, and a generic unfilled PFlink custom CSV file that users may populate with key experimental conditions and results, which can then be read and converted into the HXMS format.

Software↗

PubMed2026

Artificial Intelligence for Natural Products Discovery and Development.

Natural products (NPs) remain a cornerstone of modern drug discovery, offering stereochemical complexity and diverse bioactivities that precisely modulate therapeutic targets, refined through billions of years of evolution. However, their research has long been hindered by inefficient, empirical workflows, high resource consumption, structural complexity, and the "multicomponent, multi-target" nature of their mechanisms. The exponential growth of genomic, metabolomic, and spectral data has overwhelmed conventional analytical methods, exposing critical bottlenecks in handling high-dimensional, heterogeneous datasets that exceed human interpretive capacity. Artificial intelligence (AI) is emerging as a transformative paradigm to address these challenges, integrating multi-omics and chemical data to shift NP research from fragmented empiricism toward mechanism-driven, precision-oriented development. By leveraging deep learning architectures- including graph neural networks, Transformers, and diffusion-based generative models-AI enables systematic decoding of NP biosynthesis, automated structure elucidation, rational target identification, knowledge extraction from vast unstructured scientific literature, and de novo molecular design. This review comprehensively surveys recent advances in AI applications across the full NP discovery and development pipeline, encompassing genome mining, structure-based and ligand-based virtual screening, multimodal structural characterization, lead optimization, and biosynthetic pathway engineering. We further examine the emerging roles of protein-centric, molecule- centric, and multimodal foundation models, as well as large language models, in bridging genotype-to-chemotype gaps and unlocking unstructured scientific knowledge. Finally, we discuss critical challenges including data scarcity, representational limitations for complex stereochemistry, physical plausibility in generative models, and the urgent need for experimental validation, while outlining future directions toward autonomous experimentation, closed-loop optimization, and human-AI collaborative discovery.

Artificial intelligence↗

PubMed1997

A neural model of multimodal adaptive saccadic eye movement control by superior colliculus.

How does the saccadic movement system select a target when visual, auditory, and planned movement commands differ? How do retinal, head-centered, and motor error coordinates interact during the selection process? Recent data on superior colliculus (SC) reveal a spreading wave of activation across buildup cells the peak activity of which covaries with the current gaze error. In contrast, the locus of peak activity remains constant at burst cells, whereas their activity level decays with residual gaze error. A neural model answers these questions and simulates burst and buildup responses in visual, overlap, memory, and gap tasks. The model also simulates data on multimodal enhancement and suppression of activity in the deeper SC layers and suggests a functional role for NMDA receptors in this region. In particular, the model suggests how auditory and planned saccadic target positions become aligned and compete with visually reactive target positions to select a movement command. For this to occur, a transformation between auditory and planned head-centered representations and a retinotopic target representation is learned. Burst cells in the model generate teaching signals to the spreading wave layer. Spreading waves are produced by corollary discharges that render planned and visually reactive targets dimensionally consistent and enable them to compete for attention to generate a movement command in motor error coordinates. The attentional selection process also helps to stabilize the map-learning process. The model functionally interprets cells in the superior colliculus, frontal eye field, parietal cortex, mesencephalic reticular formation, paramedian pontine reticular formation, and substantia nigra pars reticulata.

Animals↗

PubMed2005

Motor-maps, navigation and implicit space representation in the hippocampus.

Multiple sensory-motor maps located in the brainstem and the cortex are involved in spatial orientation. Guiding movements of eyes, head, neck and arms they provide an approximately linear relation between target distance and motor response. This involves especially the superior colliculus in the brainstem and the parietal cortex. There, the natural frame of reference follows from the retinal representation of the environment. A model of navigation is presented that is based on the modulation of activity in those sensory-motor maps. The actual mechanism chosen was gain-field modulation, a process of multimodal integration that has been demonstrated in the parietal cortex and superior colliculus, and was implemented as attraction to visual cues (colour). Dependent on the metric of the sensory-motor map, the relative attraction to these cues implemented as gain field modulation and their position define a fixed point attractor on the plane for locomotive behaviour. The actual implementation used Kohonen-networks in a variant of reinforcement learning that are well suited to generate such topographically organized sensory-motor maps with roughly linear visuo-motor response characteristics. In the following, it was investigated how such an implicit coding of target positions by gain-field parameters might be represented in the hippocampus formation and under what conditions a direction-invariant space representation can arise from such retinotopic representations of multiple cues. Information about the orientation in the plane--as could be provided by head direction cells--appeared to be necessary for unambiguous space representation in our model in agreement with physiological experiments. With this information, Gauss-shaped "place-cells" could be generated, however, the representation of the spatial environment was repetitive and clustered and single cells were always tuned to the gain-field parameters as well.

Algorithms↗

PubMed1981

Vision and kinesthesis in spatial short-term memory.

This investigation examined the effects of different localizing cues on the precision with which subjects can recall the position of a target in space. The availability of the cues--vision, kinesthesis, or both--was varied during both learning and recall. Subjects (age range 26 to 58 yr.) placed the stimulus or watched it being placed. The stimulus was removed, and subjects replaced or indicated its replacement location. Results show a striking similarity of performance for all subjects and significant performance differences relative to the different cues. The results provide information about intramodal and intermodal visual and kinesthetic transfer effects and about the effects of unimodal and multimodal input in such transfer. Results confirm a dominance of vision over kinesthesis, i.e., "kinesthetic memory" does not provide as accurate localizing information as does "visual memory."

Adult↗

PubMed2006

[Tourette syndrome: from a neurological clinic to a multidisciplinary approach].

Tourette Syndrome (TS) is a chronic, familial disorder, characterized by involuntary motor and phonic tics that wax and wane in severity. TS is frequently accompanied by behavioral, emotional and cognitive problems that are often more incapacitating than the tic disorder itself. After a review of the disorder, in which the multidisciplinary aspects are emphasized, the article describes the clinical features of 60 children with TS, 49 boys and 11 girls, aged 13 +/- 3.6 years (mean SD), treated in the Neuropediatric Unit at Shaare Zedek Medical Center. The children described had both motor and vocal tics, but also had ADHD (n = 44), obsessive-compulsive disorder (n = 32), learning disabilities (with 12 children learning in special education frameworks) and behavioral disorders (n = 36). The clinical profile of this group of children with TS is similar to that reported on referred patients regardless of cultural background. Since children with TS manifest multiple comorbidities, optimal therapy mandates the cooperation of a multidisciplinary team including a pediatric neurologist, a child psychiatrist, a psychologist and a family therapist. Working in concert, these specialists can implement a multimodal approach, addressing the neurological and psychiatric aspects of TS as well as enhancing the child's coping skills with the disorder itself and its consequences.

Child↗

PubMed2026

Osteoarthritis phenotypes: advancing precision medicine through clinical, structural, and molecular stratification.

PURPOSE: Osteoarthritis (OA) is now understood as a heterogeneous syndrome driven by diverse biological, biomechanical, metabolic, genetic, and molecular mechanisms. This variability explains differences in disease progression and treatment response, challenging the traditional "one-size-fits-all" approach. This review highlights OA phenotyping as a key step toward precision medicine, focusing on clinical, structural, and molecular classifications that inform individualized care. METHODS: A narrative review was conducted using a non-systematic search of major databases and Osteoarthritis Research Society International sources (2010-2026). Evidence was thematically synthesized across clinical, imaging, and molecular domains to characterize OA phenotypes and their potential relevance to precision medicine. RESULTS: Multiple OA phenotypes were identified: inflammatory, metabolic, biomechanical, cartilage-subchondral, pain-sensitization, and aging/senescence. These exhibit distinct clinical features, risk factors, and therapeutic responses. Imaging-based phenotypes (e.g., inflammatory, meniscus-cartilage, subchondral bone, atrophic, hypertrophic) and molecular endotypes (low turnover, structural damage, systemic inflammation) further refine stratification. Pain-structure discordance is notable in sensitization phenotypes and may predict poorer surgical outcomes. Joint-specific variations and emerging genomic and epigenetic insights underscore disease complexity. Advances in imaging, biomarkers, and machine learning may enable earlier detection and patient clustering, though clinical application remains limited. CONCLUSION: Phenotype- and endotype-based classification represents a critical advancement toward precision OA management. Tailored interventions based on stratification hold promise for improving outcomes; however, clinical translation remains limited by overlapping phenotypes, lack of validated biomarkers, and inconsistent results from phenotype-driven trials. Wider clinical adoption requires standardized definitions, validation across joints, and integration of multimodal diagnostic tools into routine practice.

Humans↗

PubMed1998

Characterization of perforant path lesions in rodent models of memory and attention.

Early stage Alzheimer's disease (AD) pathology is associated with neurodegeneration of systems within the temporal cortex, e.g. the entorhinal cortex, perforant pathway and hippocampus. The perforant pathway provides the major neuronal input to the hippocampus from the entorhinal cortex and thus relays multimodal sensory information derived from cortical zones into the hippocampus. The earliest symptoms of AD include cognitive impairments, e.g. deficits in short-term memory and attention. Consequently, we have investigated the effect of bilateral knife cut lesions to the perforant path on cognition in rats using models measuring primarily short-term memory (operant delayed match to position task), attention (serial five-choice reaction time task) and spatial learning (Morris water maze). Rats receiving bilateral perforant path lesions showed normal neurological function and a mild hyperactivity. The lesion produced little effect on attention assessed using the five-choice task. In contrast, animals with equivalent lesions showed a robust delay-dependent deficit in the delayed match to position task. Spatial learning in the water maze task was also severely impaired. The delay-dependent deficit in the match to position task was not reversed by tacrine (3 mg/kg) pretreatment. The present data support a selective impairment of cognitive function following perforant path lesions that was confined to mnemonic rather than attentional processing. These findings complement primate and human studies identifying a critical role of the perforant pathway and associated temporal lobe structures in declarative memory. Degeneration of the perforant pathway is likely to contribute to the mnemonic deficits characteristic of early AD. The failure of tacrine to ameliorate these deficits may be relevant to an emerging clinical literature suggesting that cholinomimetic therapies improve attentional rather than mnemonic function in AD.

Animals↗

PubMed2005

Functional, structural, and metabolic abnormalities of the hippocampal formation in Williams syndrome.

Williams syndrome (WS), caused by microdeletion of some 21 genes on chromosome 7q11.23, is characterized by dysmorphic features, mental retardation or learning difficulties, elastin arteriopathy, and striking neurocognitive and social-behavioral abnormalities. Recent studies of murine knockouts of key genes in the microdeleted region, LIM kinase 1 (LIMK1) and cytoplasmatic linker protein 2 (CYLN2), demonstrated significant functional and metabolic abnormalities, but grossly normal structure, in the hippocampal formation (HF). Furthermore, deficits in spatial navigation and long-term memory, major cognitive domains dependent on hippocampal function, have been described in WS. We used multimodal neuroimaging to characterize hippocampal structure, function, and metabolic integrity in 12 participants with WS and 12 age-, sex-, and IQ-matched healthy controls. PET and functional MRI studies showed profound reduction in resting blood flow and absent differential response to visual stimuli in the anterior HF in WS. Spectroscopic measures of N-acetyl aspartate, considered a marker of synaptic activity, were reduced. Hippocampal size was preserved, but subtle alterations in shape were present. These data demonstrate abnormalities in HF in WS in agreement with murine models, implicate LIMK1 and CYLN2 in human hippocampal function, and suggest that hippocampal dysfunction may contribute to neurocognitive abnormalities in WS.

Adult↗

PubMed2004

Clinical applications of cognitive event-related potentials in Alzheimer's disease.

This article has reviewed several abnormalities in the cognitive ERPs of AD patients. These abnormalities are prominent from latencies of approximately 200 msec and later. In contrast, sensory-dependent evoked potentials, such as N100, are generally normal in AD. This finding is as one familiar with the neuropathology of AD would predict. Predilection sites in early AD include the medial temporal lobe, other limbic areas, and multimodal association cortices with sparing of primary sensory areas. Unimodal association cortex is involved in AD, but not as heavily as multimodal cortex. Particular advantages of studying a given ERP paradigm or component depend largely on the specific application or hypothesis being tested. A P300 paradigm can be useful in detecting a disorder of attention or in quantifying the effects of drugs that improve attention, such as the cholinesterase inhibitors. For the early diagnosis of AD or other memory disorders, a word-repetition paradigm with an explicit recognition task or one that fosters associative learning would be recommended. This article has discussed potential use of N400 in tracking disease progression. ERPs provide a flexible and powerful technique, with superb temporal resolution, which can be used as a probe into subtle "subclinical" abnormalities of cognitive processes. Despite being applied to AD for about 25 years since the early P300 studies, the full potential of ERPs in helping diagnose and treat AD patients has yet to be realized. In this era of rapidly evolving brain-imaging techniques, electrophysiologic data are important in advancing understanding of cognition. Brain-mapping techniques that can inform where and when key cognitive processes occur are finally emerging. A final example of potential clinical application of cognitive ERPs is in the development of rational combinational treatment of cognitive enhancing drugs. Along these lines, P300 investigations in epilepsy proved helpful in ranking the cognitive side effects of anticonvulsant drugs. Drug studies that use 2 x 2 combinational designs, which compare the effects of drug A, drug B, with A + B, are currently prohibitively expensive for full-scale clinical trials in AD. It is likely that precise ERP measures could hasten drug development in several ways. Smaller samples could be used, at lower cost, to test the cognitive effects of each specific drug combination. Optimal doses of combinational therapy perhaps could be identified by repeated within-subject ERP measures. Longitudinal changes in the ERP hold promise as a marker of individual responsivity to a particular agent, which could have diagnostic utility (eg, testing response to cholinergic or dopaminergic therapy). This horizon and many others remain wide open for well-planned explorations.

Aging↗

PubMed1979

The effect of delayed feedback on infant learning reexamined.

The study reexamines the effect of delayed reinforcement upon contingency behavior in 6- to 8-month-old infants and attempts to account for the temporal discrepancy between span of integration and contingency memory. A modified delayed-reinforcement scheduling procedure enabled a previous methodological criticism to be discounted. The findings confirmed that whereas infants revealed reliable acquisition under immediate reinforcement, a 3-sec delay (whether reset or nonreset) precluded response acquisition, as did 6-sec and 10-sec delay of reinforcement. The findings are interpreted in terms of an informational-load hypothesis which relates short-term memory to the integration and/or segregation of multimodal input.

Conditioning, Operant↗

PubMed1998

Subclinical cerebral complications after coronary artery bypass grafting: prospective analysis with magnetic resonance imaging, quantitative electroencephalography, and neuropsychological assessment.

OBJECTIVE: To analyze the frequency and severity of subclinical cerebral complications associated with coronary artery bypass grafting (CABG). DESIGN: A prospective controlled study using preoperative and postoperative magnetic resonance imaging (MRI) of the brain, quantitative electroencephalography (QEEG), and detailed neuropsychological and neurologic examinations as potentially sensitive indicators of subclinical cerebral injury associated with CABG. SETTING: Multimodality evaluation in a tertiary care unit (Kuopio University Hospital, Kuopio, Finland). PATIENTS: Thirty-eight patients undergoing elective CABG and 20 control patients undergoing other major vascular surgery, mostly operations on the abdominal aorta. MAIN OUTCOME MEASURES: Coronary artery bypass grafting-associated cerebral complications assessed preoperatively and postoperatively by brain MRI, QEEG, detailed neurologic examination, and a neuropsychological test battery that evaluates cognitive functions in major areas known to be vulnerable to organic impairment (learning and memory, attention, flexible mental processing, and psychomotor speed). RESULTS: There were no major neurologic complications. A mild hemisyndrome developed in 1 patient who underwent CABG and in 1 control patient. Overall, there was no decline in mean cognitive performance 3 months after surgery. Electroencephalographic slowing of 0.5 Hz or more in at least 2 channels occurred in 11 patients who underwent CABG and in 1 control patient (P=.03). The postoperative brain MRI scan revealed new small ischemic lesions in 8 patients (21%) in the CABG group but in none of the control group (P=.03). These new cerebral MRI lesions did not explain deterioration in neuropsychological test performance or the QEEG slowing. CONCLUSIONS: Coronary artery bypass grafting causes more QEEG alterations and small ischemic cerebral lesions that are detectable by MRI than does other major vascular surgery. The effect is mainly subclinical, because no statistically significant deterioration in mean neuropsychological test performance was detected.

Aged↗

PubMed2002

Engineering and algorithm design for an image processing Api: a technical report on ITK--the Insight Toolkit.

We present the detailed planning and execution of the Insight Toolkit (ITK), an application programmers interface (API) for the segmentation and registration of medical image data. This public resource has been developed through the NLM Visible Human Project, and is in beta test as an open-source software offering under cost-free licensing. The toolkit concentrates on 3D medical data segmentation and registration algorithms, multimodal and multiresolution capabilities, and portable platform independent support for Windows, Linux/Unix systems. This toolkit was built using current practices in software engineering. Specifically, we embraced the concept of generic programming during the development of these tools, working extensively with C++ templates and the freedom and flexibility they allow. Software development tools for distributed consortium-based code development have been created and are also publicly available. We discuss our assumptions, design decisions, and some lessons learned.

Algorithms↗

PubMed1987

Transfer of sensory-integration training.

96 university students participated in one of six treatments to assess whether training to make cross-modal or intramodal matches of one linear-movement distance would transfer to other distances. Three groups received unimodal training (vision, audition, or kinesthesis) and three groups received multimodal training (all combinations of vision, audition, and kinesthesis) to make kinesthetic matches of one common linear-movement distance. All subjects were subsequently tested on kinesthetic matches of novel distances both shorter and longer than the training distance. A multivariate analysis of variance and Bonferroni contrasts indicated that cross-modal matching was enhanced by multimodal experience but only for distances shorter than the training distance. It was speculated that transferring to distances shorter than a training distance might involve processes different from those used to transfer to distances longer than the training distance.

Adolescent↗

PubMed2000

[Selective interference, process dissociation and serial information in recall of performed actions. A comment on Thorsten Meiser's remarks on the multimodal memory theory].

In this commentary, I deal with three questions. (1) Do the findings of interference experiments justify the distinction between motor and visual-imaginal processes? (2) Can automatic and controlled process components be identified by the process-dissociation technique? (3) Does encoding of order information depend on encoding instructions?

Attention↗

PubMed2026

A comparative evaluation of multiple enlarged perivascular space segmentation tools.

BACKGROUND: Enlarged perivascular spaces (ePVS) are a marker of cerebral small vessel disease, potentially reflecting reduced waste clearance. Because manual quantification is unfeasible in large datasets, we developed and evaluated an automated tool. METHODS: Detection Of Regions of Enlarged perivascular Spaces (DORES), a 3D nnU-Net-based deep learning algorithm was developed for ePVS segmentation using T1-weighted and fluid-attenuated inversion recovery magnetic resonance imaging (MRI). DORES was developed in two stages: an initial model trained on 35 manually segmented scans and a final model on 1460 pseudo-labeled sessions from the Vanderbilt Memory and Aging Project (VMAP). A subset of VMAP participants with 3 T brain MRI underwent whole-brain manual ePVS tracing (n = 35, 73 ± 9 years, 51% male) and visual rating (n = 388, 71 ± 8 years, 54% male) by a neuroradiologist. DORES was evaluated and compared against three other segmentation tools using Dice and F1 scores, absolute volume and element differences, correlation, and agreement. External validation used an Alzheimer's Disease Neuroimaging Initiative 3 subset with manual tracings (ADNI3, n = 18, 73 ± 9 years, 67% female). RESULTS: DORES achieved Dice scores of 0.61 ± 0.16 (white matter) and 0.72 ± 0.08 (basal ganglia) in VMAP, with strong correlations and agreement for ePVS count and volume. Performances modestly declined in ADNI3 across algorithms. Scanner-stratified analyses showed stronger correlations for Philips versus Siemens images in the basal ganglia, indicating scanner-dependent differences in measurement consistency. CONCLUSIONS: DORES provides a multimodal nnU-Net-based pipeline for ePVS segmentation in older adults. The model demonstrates robust within-cohort performance and reasonable external validity, though scanner-related effects limit application across sites.

Humans↗