PubMed Health⌕ Search

SEARCH · PubMed Health

Results for “Large Language 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 343 records · Page 19Linked to original sources

Brain imaging studies of the anatomical and functional consequences of preterm birth for human brain development.

Premature birth can have devastating effects on brain development and long-term functional outcome. Rates of psychiatric illness and learning difficulties are high, and intelligence on average is lower than population means. Brain imaging studies of infants born prematurely have demonstrated reduced volumes of parietal and sensorimotor cortical gray matter regions. Studies of school-aged children have demonstrated reduced volumes of these same regions, as well as in temporal and premotor regions, in both gray and white matter. The degrees of these anatomical abnormalities have been shown to correlate with cognitive outcome and with the degree of fetal immaturity at birth. Functional imaging studies have shown that these anatomical abnormalities are associated with severe disturbances in the organization and use of neural systems subserving language, particularly for school-aged children who have low verbal IQs. Animal models suggest that hypoxia-ischemia may be responsible at least in part for some of the anatomical and functional abnormalities. Increasing evidence suggests that a host of mediators for hypoxic-ischemic insults likely contribute to the disturbances in brain development in preterm infants, including increased apoptosis, free-radical formation, glutamatergic excitotoxicity, and alterations in the expression of a large number of genes that regulate brain maturation, particularly those involved in the development of postsynaptic neurons and the stabilization of synapses. The collaboration of both basic neuroscientists and clinical researchers is needed to understand how normal brain development is derailed by preterm birth and to develop effective prevention and early interventions for these often devastating conditions.

Brain↗

An existential model of oral health from evolving views on health, function and disability.

OBJECTIVE: This study explores the evolution of conceptual frameworks and models of health and disability to construct an explanatory model of oral health. RESULTS: The International Classification of Impairments, Disabilities, and Handicaps adopted by the WHO is based largely on social role theory and a utilitarian tradition portraying disablement as a negative and socially unacceptable consequence of impairment. It has been the major conceptual influence on the construction of psychometric tools for dentistry. However current views of chronic disease are refocused on the influence of coping strategies used by people to prevent or limit disability and handicap. Consequently, the WHO adopted the International Classification of Functioning, Disability and Health (ICF) as an alternative description of health and health-related states based on an existentialist view of the body, the person and society. In addition, an ethnographic exploration has identified three major domains of oral health--oral hygiene, comfort and general health--that dominate the opinions of people with oral impairments. CONCLUSIONS: Application of the framework and language of the ICF to the major domains of oral health provides the basis for a new biopsychosocial model of oral health, function and disablement.

Disability Evaluation↗

The community residential treatment service: developing a continuum of prosthetic environments for the chronically disabled.

Among the many issues regarding the care of chronic mental patients, none is more pressing than the need for administrative and clinical models designed to organize and systematize the efforts of diverse community service providers. This paper describes the functioning of the Community Residential Treatment Service of the South Beach Psychiatric Center, a large-scale project of a state facility created to respond to this tissue. By blending sophisticated clinical and administrative technology, programs operated by the state, voluntary, and proprietary health care sectors have been integrated to form a balanced service delivery system. This system provides a broad continuum of inpatient and outpatient residential settings developed in accordance with social learning principles. The components of the system, with the Community Residential Treatment Service as the major integrative force, are linked together by detailed contracts as well as common behavioral clinical and behavioral administrative language. The treatment successes f this system have been significant enough to suggest that a positive synergistic effect is generated by this programming combination.

Community Mental Health Services↗

Automatic sign language analysis: a survey and the future beyond lexical meaning.

Research in automatic analysis of sign language has largely focused on recognizing the lexical (or citation) form of sign gestures as they appear in continuous signing, and developing algorithms that scale well to large vocabularies. However, successful recognition of lexical signs is not sufficient for a full understanding of sign language communication. Nonmanual signals and grammatical processes which result in systematic variations in sign appearance are integral aspects of this communication but have received comparatively little attention in the literature. In this survey, we examine data acquisition, feature extraction and classification methods employed for the analysis of sign language gestures. These are discussed with respect to issues such as modeling transitions between signs in continuous signing, modeling inflectional processes, signer independence, and adaptation. We further examine works that attempt to analyze nonmanual signals and discuss issues related to integrating these with (hand) sign gestures. We also discuss the overall progress toward a true test of sign recognition systems--dealing with natural signing by native signers. We suggest some future directions for this research and also point to contributions it can make to other fields of research. Web-based supplemental materials (appendicies) which contain several illustrative examples and videos of signing can be found at www.computer.org/publications/dlib.

Algorithms↗

Functional genetic analysis of mutations implicated in a human speech and language disorder.

Mutations in the FOXP2 gene cause a severe communication disorder involving speech deficits (developmental verbal dyspraxia), accompanied by wide-ranging impairments in expressive and receptive language. The protein encoded by FOXP2 belongs to a divergent subgroup of forkhead-box transcription factors, with a distinctive DNA-binding domain and motifs that mediate hetero- and homodimerization. Here we report the first direct functional genetic investigation of missense and nonsense mutations in FOXP2 using human cell-lines, including a well-established neuronal model system. We focused on three unusual FOXP2 coding variants, uniquely identified in cases of verbal dyspraxia, assessing expression, subcellular localization, DNA-binding and transactivation properties. Analysis of the R553H forkhead-box substitution, found in all affected members of a large three-generation family, indicated that it severely affects FOXP2 function, chiefly by disrupting nuclear localization and DNA-binding properties. The R328X truncation mutation, segregating with speech/language disorder in a second family, yields an unstable, predominantly cytoplasmic product that lacks transactivation capacity. A third coding variant (Q17L) observed in a single affected child did not have any detectable functional effect in the present study. In addition, we used the same systems to explore the properties of different isoforms of FOXP2, resulting from alternative splicing in human brain. Notably, one such isoform, FOXP2.10+, contains dimerization domains, but no DNA-binding domain, and displayed increased cytoplasmic localization, coupled with aggresome formation. We hypothesize that expression of alternative isoforms of FOXP2 may provide mechanisms for post-translational regulation of transcription factor function.

Alternative Splicing↗

Donuts, scratches and blanks: robust model-based segmentation of microarray images.

MOTIVATION: Inner holes, artifacts and blank spots are common in microarray images, but current image analysis methods do not pay them enough attention. We propose a new robust model-based method for processing microarray images so as to estimate foreground and background intensities. The method starts with a very simple but effective automatic gridding method, and then proceeds in two steps. The first step applies model-based clustering to the distribution of pixel intensities, using the Bayesian Information Criterion (BIC) to choose the number of groups up to a maximum of three. The second step is spatial, finding the large spatially connected components in each cluster of pixels. The method thus combines the strengths of the histogram-based and spatial approaches. It deals effectively with inner holes in spots and with artifacts. It also provides a formal inferential basis for deciding when the spot is blank, namely when the BIC favors one group over two or three. RESULTS: We apply our methods for gridding and segmentation to cDNA microarray images from an HIV infection experiment. In these experiments, our method had better stability across replicates than a fixed-circle segmentation method or the seeded region growing method in the SPOT software, without introducing noticeable bias when estimating the intensities of differentially expressed genes. AVAILABILITY: spotSegmentation, an R language package implementing both the gridding and segmentation methods is available through the Bioconductor project (http://www.bioconductor.org). The segmentation method requires the contributed R package MCLUST for model-based clustering (http://cran.us.r-project.org). CONTACT: fraley@stat.washington.edu.

Algorithms↗

Diagnostic labels applied to model case histories of chronic airflow obstruction. Responses to a questionnaire in 11 North American and Western European countries.

In Canada, USA and 9 Western European Countries, 121 respiratory physicians responded to an English language questionnaire asking them to state how they would investigate, treat and label four model patients, chosen to represent well-recognized patterns of clinical features of chronic airflow obstruction. Selection of further investigations appeared to be determined more by the probable diagnostic label than by the need to define selected characteristics in the whole range of such patients. Differences in recommended treatment between countries were less than others have reported for the treatment of asthma. Analysis of the diagnostic labels showed: the classic terms asthma, chronic bronchitis, emphysema still predominated in clinical practice and were considered to be better defined entities than any of the many terms introduced to describe chronic airflow obstruction in the last 30 yrs; the term chronic bronchitis was a source of confusion unless qualified to indicate presence or absence of obstruction; the use of combination terms such as chronic asthmatic bronchitis and chronic obstructive bronchitis showed large differences between countries; there were few differences related to national language. The implications of these findings are discussed.

Aged↗

Enhancement of the signal-to-noise ratio in H2(15)O bolus PET activation images: a combined cold-bolus, switched protocol.

UNLABELLED: To increase the signal-to-noise ratio (S/N) of H2(15)O bolus PET activation images, we designed and tested a data acquisition protocol that alters the relative distribution of tracer in the uptake and washout phases of the input function. This protocol enhances the S/N gains obtained with conventional switched protocols by combining task switching and the use of a large bolus of blood free of tracer (cold bolus). The cold bolus is formed by sequestering blood in the lower limbs with a double cuff before tracer injection. METHODS: The effect of a combined cold-bolus, switched protocol on the signal from activation images was first simulated using a compartmental model of the uptake of H2(15)O into the brain. Then, the effectiveness of the protocol was investigated in 4 healthy volunteers performing a language task. Each volunteer underwent scanning 12 times: 3 activation/ baseline and 3 baseline/activation scans using the conventional switched protocol and 3 activation/baseline and 3 baseline/activation scans using the combined cold-bolus, switched protocol. The S/N changes introduced when using the cold bolus were analyzed by comparing, across protocols, the magnitude and statistical significance of the activation foci associated with the execution of the language task identified in the averaged subtracted images, and by comparing image noise levels. RESULTS: In the simulated datasets, the combined protocol yielded a substantial increase in the activation signals for scan durations greater than 60 s, in comparison with equivalent signals yielded by the switched protocol alone. In the PET experiments, activation foci obtained using the combined protocol had significantly higher t statistic values than did equivalent foci detected using the conventional switched protocol (mean improvement, 36%). Analysis of the S/N in the averaged subtracted images revealed that the improvements in statistical significance of the activation foci were caused by increases in the signal magnitudes and not by decreases in overall image noise. CONCLUSION: We designed a data acquisition protocol for H2(15)O bolus PET activation studies that combines the use of a tracer-free bolus with a switched protocol. Simulated and experimental data suggest that this combined protocol enhances the S/N gains obtained with a conventional switched protocol. Implementation of the combined protocol in H2(15)O bolus activation studies was easy.

Adult↗

Hemispheric asymmetries in motor function: I. Left-hemisphere specialization for memory but not performance.

Patients with unilateral brain lesions of vascular origin were administered tests designed to determine if left-hemisphere specialization in manual-sequence tasks involves memory for these sequences, or performance of them, or both. Patients with left-sided lesions were worse than patients with right-sided lesions on two tasks requiring the recall of hand positions. Whereas patients with left-sided lesions showed a trend towards being worse on speeded performance of an already learned manual sequence, in both groups on this task there were a large number of failures to remember the sequence. When memory demands were better controlled by providing a model during the speeded performance task, there were no group differences. It is proposed that there is left-hemisphere specialization for memory but not performance of such motor tasks.

Brain↗

Dynamical systems and cognitive linguistics: toward an active morphodynamical semantics.

We propose a novel dynamical system approach to cognitive linguistics based on cellular automata and spiking neural networks. How can the same relationship 'in' apply to containers as different as 'box', 'tree' or 'bowl'? Our objective is to categorize the infinite diversity of schematic visual scenes into a small set of grammatical elements and elucidate the topology of language. Gestalt-inspired semantic studies have shown that spatial prepositions such as 'in' or 'above' are neutral toward the shape and size of objects. We suggest that this invariance can be explained by introducing morphodynamical transforms, which erase image details and create virtual structures or singularities (boundaries, skeleton), and call this paradigm 'active semantics'. Singularities arise from a large-scale lattice of coupled excitable units exhibiting spatiotemporal pattern formation, in particular traveling waves. This work addresses the crucial cognitive mechanisms of spatial schematization and categorization at the interface between vision and language and anchors them to expansion processes such as activity diffusion or wave propagation.

Algorithms↗

Child health services in transition: I. Theories, methods and launching.

AIM: To describe an evidence-based model for preventive child health care and present some findings from baseline measurements. METHODS: The model includes: parent education; methods for interaction and language training; follow-up of low birthweight children; identification and treatment of postnatal depression, interaction difficulties, motor problems, parenthood stress, and psychosocial problems. After baseline measurements at 18 mo (cohort I), the intervention was tested on children from 0 to 18 mo at 18 child health centres in Uppsala County (cohort II). Eighteen centres in other counties served as controls. Two centres from a privileged area were included in the baseline measurements as a "contrasting" sample. Data are derived from health records and questionnaires to nurses and mothers. RESULTS: Baseline experiment (n = 457) and control mothers (n = 510) were largely comparable in a number of respects. Experiment parents were of higher educational and occupational status, and were more frequently of non-Nordic ethnicity. Mothers in the privileged area (n = 72) differed from other mothers in several respects. Experiment nurses devoted considerably fewer hours per week to child health services and to child patients than did control nurses. CONCLUSIONS: Despite certain differences, experiment and control samples appeared comparable enough to permit, in a second step, conclusions about the effectiveness of the intervention.

Child Health Services↗

Promoting culturally appropriate colorectal cancer screening through a health educator: a randomized controlled trial.

BACKGROUND: Colorectal cancer (CRC) is a leading cause of cancer mortality in the US. Surveys reveal low CRC screening levels among Asians in the US, including Chinese Americans. METHODS: A randomized controlled trial was conducted with Chinese patients to evaluate a clinic-based, culturally and linguistically appropriate intervention promoting fecal occult blood test (FOBT) screening. The multifaceted intervention included a trilingual and bicultural health educator, bilingual materials (a video, a motivational pamphlet, an informational pamphlet, and FOBT instructions), and three FOBT cards. Patients in the control arm received usual care. Our primary outcome measure was FOBT screening within 6 months after randomization. The proportion of FOBT completion in the intervention and control arms was compared by using a chi-square test, and logistic regression analysis was performed to adjust for the effects of sociodemographic variables and prior screening history. Potential effect modifications were also tested by using logistic regression models. RESULTS: Our intervention had a strong effect on FOBT completion (intervention group, 69.5%; control group, 27.6%), and the adjusted odds of FOBT slightly increased to over 6-fold greater in the intervention arm compared with the control arm. No effect modification by age, gender, language, insurance, or prior FOBT was found. CONCLUSIONS: The authors' multifaceted, culturally appropriate intervention significantly increased FOBT screening in a group of low-income and less-acculturated minority patients. Given the large effect size, future research should determine the effective core component(s) that can increase CRC screening in both the general and minority populations.

Aged↗

Errors in a nonlinear graphic-semantic mapping task resulting from lesions in Boltzmann machine: is it relevant to dyslexia?

One of the most fascinating aspects of brain research is the subject of language. As in many other cases, the malfunctions that occur in different persons for various reasons give us insight on the mechanisms that support our ability to talk, read and listen. Following the work of Plaut and associates, we deal with the dyslexia disorder, which is the overall name for a large number of reading disorders. A Boltzmann machine neural network scheme was trained to implement the nonlinear mapping task of graphic representation into semantic representation, which may model the brain sections responsible for the translation of a written word into meanings and syllables. After training, various types of lesions were applied and the performance of the network was tested in order to measure the effect of each lesion on the error rate and type distribution that were detected. The system's errors were classified into several categories and the distribution of errors between the categories was studied. Using the simulations, it is demonstrated that a finite scheduling process in the Boltzmann machine causes the distribution of the network's errors to be unique and different from its expected error distribution. The phenomenon is given a mathematical explanation rooted in the statistical mechanics basics of the Boltzmann machine. Test results suggest the localization of certain reading functions within the network. Comparison is made to relevant types of dyslexia and shows resemblance in major symptoms as well as in certain known side effects.

Algorithms↗

Nursing process outcome linkage research: issues, current status, and health policy implications.

BACKGROUND: The use of large clinical datasets to assess the effectiveness of health care is of growing interest in continuing efforts to understand the impact of healthcare costs on quality. Correspondingly, there is a greater need to define and measure outcomes that are sensitive to nursing interventions. However, concerns exist about the ability to amass and use large clinical nursing datasets to assess the effectiveness of nursing interventions. Some nursing studies have used large clinical datasets to examine patterns of nursing diagnoses, interventions, and outcomes. Among patient populations, however, systematic effectiveness studies of nursing process and outcome linkages at the individual nurse and patient level of analysis are essentially nonexistent. This is largely the result of slow development of nursing classifications, reference terminologies, and reference information standards. Nursing information systems have an unprecedented potential for documentation of nursing practice, as well as the accumulation and analysis of large clinical datasets, to improve nursing performance, increase nursing knowledge, and provide data and information necessary for nursing to participate in the formulation of healthcare policy. OBJECTIVES: A literature search shows that a common framework is beginning to evolve that represents nursing's essential information, eg, the Nursing Minimum Data Set, Management Minimum Data Set, and several standardized nursing languages. Extensive research and other initiatives have produced 1) nursing languages and reference terminologies that span healthcare settings; 2) information models; and 3) standards for datasets supporting information systems. A number of issues remain, however, that concern the development of uniform nursing datasets, definitions of outcomes, quality of nursing data, information system design, and methods of data analysis. We review nursing process outcome research, clarify issues inherent in nursing effectiveness research, and discuss implications for nursing and health policy.

Health Policy↗

State of the art in continuous speech recognition.

In the past decade, tremendous advances in the state of the art of automatic speech recognition by machine have taken place. A reduction in the word error rate by more than a factor of 5 and an increase in recognition speeds by several orders of magnitude (brought about by a combination of faster recognition search algorithms and more powerful computers), have combined to make high-accuracy, speaker-independent, continuous speech recognition for large vocabularies possible in real time, on off-the-shelf workstations, without the aid of special hardware. These advances promise to make speech recognition technology readily available to the general public. This paper focuses on the speech recognition advances made through better speech modeling techniques, chiefly through more accurate mathematical modeling of speech sounds.

Algorithms↗

DARKIN: a zero-shot benchmark for phosphosite-dark kinase association using protein language models.

MOTIVATION: Protein language models (pLMs) have emerged as powerful tools for capturing the intricate information encoded in protein sequences, facilitating various downstream protein prediction tasks. With numerous pLMs available, there is a critical need for diverse benchmarks to systematically evaluate their performance across biologically relevant tasks. Here, we introduce DARKIN, a zero-shot classification benchmark designed to assign phosphosites to understudied kinases, termed dark kinases. Kinases, which catalyze phosphorylation, are central to cellular signaling pathways. While phosphoproteomics enables the large-scale identification of phosphosites, determining the cognate kinase responsible for the phosphorylation event remains an experimental challenge. RESULTS: In DARKIN, we prepared training, validation, and test folds that respect the zero-shot nature of this classification problem, incorporating stratification based on kinase groups and sequence similarity. We evaluated multiple pLMs using two zero-shot classifiers: a novel, training-free k-NN-based method, and a bilinear classifier. Our findings indicate that ESM, ProtT5-XL, and SaProt exhibit superior performance on this task. DARKIN provides a challenging benchmark for assessing pLM efficacy and fosters deeper exploration of under-characterized (dark) kinases by offering a biologically relevant test bed. AVAILABILITY AND IMPLEMENTATION: The DARKIN benchmark data and the scripts for generating additional splits are publicly available at: https://github.com/tastanlab/darkin.

Protein Kinases↗

Magnetoencephalography in presurgical epilepsy evaluation.

The introduction of whole-head magnetoencephalography (MEG) systems facilitating simultaneous recording from the entire brain surface has led to a major breakthrough of MEG in presurgical epilepsy evaluation. Localizations of the interictal spike zone with MEG showed excellent agreement with invasive electrical recordings, were useful to clarify the spatial relationship of the irritative spike zone to structural lesions, and could attribute epileptic activity to lobar subcompartments both in temporal lobe and extratemporal epilepsy. MEG was especially useful for the study of patients with non-lesional neocortical epilepsy and of patients with large lesions, where it provided unique information on the epileptogenic zone. It could reliably localize sensorimotor cortex prior to surgical procedures adjacent to central fissure. MEG language mapping yielded concordant results with the Wada test and cortical stimulation studies. MEG localizations of epileptic activity and essential brain regions were successfully integrated into frameless stereotaxy systems providing accurate functional information intraoperatively. Because MEG and EEG yield both complementary and confirmatory information, combined MEG-EEG recordings in conjunction with advanced source modeling techniques will further improve the noninvasive evaluation of epilepsy patients and constantly reduce the need for invasive procedures.

Brain↗

STEPP--Search Tool for Exploration of Petri net Paths: a new tool for Petri net-based path analysis in biochemical networks.

To understand biochemical processes caused by, e. g., mutations or deletions in the genome, the knowledge of possible alternative paths between two arbitrary chemical compounds is of increasing interest for biotechnology, pharmacology, medicine, and drug design. With the steadily increasing amount of data from high-throughput experiments new biochemical networks can be constructed and existing ones can be extended, which results in many large metabolic, signal transduction, and gene regulatory networks. The search for alternative paths within these complex and large networks can provide a huge amount of solutions, which can not be handled manually. Moreover, not all of the alternative paths are generally of interest. Therefore, we have developed and implemented a method, which allows us to define constraints to reduce the set of all structurally possible paths to the truly interesting path set. The paper describes the search algorithm and the constraints definition language. We give examples for path searches using this dedicated special language for a Petri net model of the sucrose-to-starch breakdown in the potato tuber.

Algorithms↗