Characterizing phonemic fluency by transfer learning with deep language models
Ontology highlight
ABSTRACT:
SUBMITTER:
PROVIDER: S-EPMC10691875 | biostudies-literature | 2023 Jan
REPOSITORIES: biostudies-literature
Similar Datasets
Project description:The voluntary generation of non-overlearned responses is usually assessed with phonemic fluency. Like most frontal tasks, it draws upon different complex processes and systems whose precise nature is still incompletely understood. Many claimed aspects regarding the pattern of phonemic fluency performance and its underlying anatomy remain controversial. Major limitations of past investigations include small sample size, scant analysis of phonemic output and methodologically insufficient lesion analysis approaches. We investigated a large number of patients with focal unilateral right or left frontal (n = 110) or posterior (n = 100) or subcortical (n = 65) lesions imaged with magnetic resonance or computed tomography and compared their performance on the number of overall responses, words produced over time, extremely infrequent/unknown words and inappropriate words generated. We also employed, for the first time parcel-based lesion-symptom mapping, tract-wise statistical analysis as well as Bayesian multi-variate analysis based on meta-analytically defined functional region of interest, including their interactions. We found that left frontal damage was associated with greater impairment than right frontal or posterior damage on overall fluency performance, suggesting that phonemic fluency shows specificity to frontal lesions. We also found that subcorticals, similar to frontals, performed significantly worse than posteriors on overall performance suggesting that subcortical regions are also involved. However, only frontal effects were found for words produced over time, extremely infrequent/unknown and inappropriate words. Parcel-based lesion-symptom mapping analysis found that worse fluency performance was associated with damage to the posterior segment of the left frontal middle and superior gyrus, the left dorsal anterior cingulate gyrus and caudate nucleus. Tract-wise statistical analysis revealed that disconnections of left frontal tracts are critical. Bayesian multi-variate models of lesions and disconnectome maps implicated left middle and inferior frontal and left dorsomedial frontal regions. Our study suggests that a set of well localized left frontal areas together with subcortical regions and several left frontal tracts are critical for word generation. We speculate that a left lateralized network exists. It involves medial, frontal regions supporting the process of 'energization', which sustains activation for the duration of the task and middle and inferior frontal regions concerned with 'selection', required due to the competition produced by associated stored words, respectively. The methodology adopted represents a promising and empirically robust approach in furthering our understanding of the neurocognitive architecture underpinning executive processes.
| S-EPMC8530259 | biostudies-literature
Project description:Many macromolecules in biological systems exist in the form of helical polymers. However, the inherent polymorphism and heterogeneity of samples complicate the reconstruction of helical polymers from cryo-EM images. Currently, available 2D classification methods are effective at separating particles of interest from contaminants, but they do not effectively differentiate between polymorphs, resulting in heterogeneity in the 2D classes. As such, it is crucial to develop a method that can computationally divide a dataset of polymorphic helical structures into homogenous subsets. In this work, we utilized deep-learning language models to embed the filaments as vectors in hyperspace and group them into clusters. Tests with both simulated and experimental datasets have demonstrated that our method - HLM (Helical classification with Language Model) can effectively distinguish different types of filaments, in the presence of many contaminants and low signal-to-noise ratios. We also demonstrate that HLM can isolate homogeneous subsets of particles from a publicly available dataset, resulting in the discovery of a previously unreported filament variant with an extra density around the tau filaments.
| S-EPMC10843845 | biostudies-literature
Project description:Neural compensatory mechanisms associated with broad cognitive abilities have been studied. However, those associated with specific cognitive subdomains (e.g., verbal fluency) remain to be investigated in healthy aging. Here, we delineate: (a) neural substrates of verbal (phonemic) fluency, and (b) compensatory mechanisms mediating the association between these neural substrates and phonemic fluency. We analyzed resting-state functional magnetic resonance imaging from 133 right-handed, cognitively normal individuals who underwent the Controlled Oral Word Association Test (COWAT) to record their phonemic fluency. We evaluated functional connectivity in an established and extended language network comprising Wernicke, Broca, thalamic and anti-correlated modules. (a) We conducted voxel-wise multiple linear regression to identify the brain areas associated with phonemic fluency. (b) We used mediation effects of cognitive reserve, measured by the Wechsler Adult Intelligence Scale-Information subtest, upon the association between functional connectivity and phonemic fluency tested to investigate compensation. We found that: (a) Greater functional connectivity between the Wernicke module and brain areas within the anti-correlated module was associated with better performance in phonemic fluency, (b) Cognitive reserve was an unlikely mediator in younger adults. In contrast, cognitive reserve was a partial mediator of the association between functional connectivity and phonemic fluency in older adults, likely representing compensation to counter the effect of aging. We conclude that in healthy aging, higher performance in phonemic fluency at older ages could be attributed to greater functional connectivity partially facilitated by higher cognitive reserve, presumably reflecting compensatory mechanisms to minimize the effect of aging.
| S-EPMC8287584 | biostudies-literature
Project description:Geometric deep learning has recently achieved great success in non-Euclidean domains, and learning on 3D structures of large biomolecules is emerging as a distinct research area. However, its efficacy is largely constrained due to the limited quantity of structural data. Meanwhile, protein language models trained on substantial 1D sequences have shown burgeoning capabilities with scale in a broad range of applications. Several preceding studies consider combining these different protein modalities to promote the representation power of geometric neural networks but fail to present a comprehensive understanding of their benefits. In this work, we integrate the knowledge learned by well-trained protein language models into several state-of-the-art geometric networks and evaluate a variety of protein representation learning benchmarks, including protein-protein interface prediction, model quality assessment, protein-protein rigid-body docking, and binding affinity prediction. Our findings show an overall improvement of 20% over baselines. Strong evidence indicates that the incorporation of protein language models' knowledge enhances geometric networks' capacity by a significant margin and can be generalized to complex tasks.
| S-EPMC10457366 | biostudies-literature
Project description:MotivationAntimicrobial peptides (AMPs) are increasingly being used in the development of new therapeutic drugs in areas such as cancer therapy and hypertension. Additionally, they are seen as an alternative to antibiotics due to the increasing occurrence of bacterial resistance. Wet-laboratory experimental identification, however, is both time-consuming and costly, so in silico models are now commonly used in order to screen new AMP candidates.ResultsThis paper proposes a novel approach for creating model inputs; using pre-trained language models to produce contextualized embeddings, representing the amino acids within each peptide sequence, before a convolutional neural network is trained as the classifier. The results were validated on two datasets-one previously used in AMP prediction research, and a larger independent dataset created by this paper. Predictive accuracies of 93.33% and 88.26% were achieved, respectively, outperforming previous state-of-the-art classification models.Availability and implementationAll codes are available and can be accessed here: https://github.com/williamdee1/LMPred_AMP_Prediction.Supplementary informationSupplementary data are available at Bioinformatics Advances online.
| S-EPMC9710646 | biostudies-literature
Project description:Oxidation states (OS) are the charges on atoms due to electrons gained or lost upon applying an ionic approximation to their bonds. As a fundamental property, OS has been widely used in charge-neutrality verification, crystal structure determination, and reaction estimation. Currently, only heuristic rules exist for guessing the oxidation states of a given compound with many exceptions. Recent work has developed machine learning models based on heuristic structural features for predicting the oxidation states of metal ions. However, composition-based oxidation state prediction still remains elusive so far, which has significant implications for the discovery of new materials for which the structures have not been determined. This work proposes a novel deep learning-based BERT transformer language model BERTOS for predicting the oxidation states for all elements of inorganic compounds given only their chemical composition. This model achieves 96.82% accuracy for all-element oxidation states prediction benchmarked on the cleaned ICSD dataset and achieves 97.61% accuracy for oxide materials. It is also demonstrated how it can be used to conduct large-scale screening of hypothetical material compositions for materials discovery.
| S-EPMC10558692 | biostudies-literature
Project description:Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image-language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP's accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P < 0.05). For patients with referral DR, those in the PCP+DeepDR-LLM arm were more likely to adhere to DR referrals (P < 0.01). Additionally, DeepDR-LLM deployment improved the quality and empathy level of management recommendations. Given its multifaceted performance, DeepDR-LLM holds promise as a digital solution for enhancing primary diabetes care and DR screening.
| S-EPMC11485246 | biostudies-literature
Project description:Verbal fluency is the ability to retrieve lexical knowledge quickly and efficiently and develops during childhood and adolescence. Few studies have investigated associations between verbal fluency performance and brain structural variation in children. Here we examined associations of verbal fluency performance with structural measures of frontal and temporal language-related brain regions and their connections in 73 typically-developing children aged 7-13 years. Tract-based spatial statistics was used to extract fractional anisotropy (FA) from the superior longitudinal fasciculus/arcuate fasciculus (SLF/AF), and the white matter underlying frontal and temporal language-related regions. FreeSurfer was used to extract cortical thickness and surface area. Better semantic and phonemic fluency performance was associated with higher right SLF/AF FA, and phonemic fluency was also modestly associated with lower left SLF/AF FA. Explorative voxelwise analyses for semantic fluency suggested associations with FA in other fiber tracts, including corpus callosum and right inferior fronto-occipital fasciculus. Overall, our results suggest that verbal fluency performance in children may rely on right hemisphere structures, possibly involving both language and executive function networks, and less on solely left hemisphere structures as often is observed in adults. Longitudinal studies are needed to clarify whether these associations are mediated by maturational processes, stable characteristics and/or experience.
| S-EPMC8242963 | biostudies-literature
Project description:Phonemic and semantic fluency are neuropsychological tests widely used to assess patients' language and executive abilities and are highly sensitive tests in detecting language deficits in glioma patients. However, the networks that are involved in these tasks could be distinct and suggesting either a frontal (phonemic) or temporal (semantic) involvement. 42 right-handed patients (26 male, mean age = 52.5 years, SD=±13.3) were included in this retrospective study. Patients underwent awake (54.8%) or asleep (45.2%) surgery for low-grade (16.7%) or high-grade-glioma (83.3%) in the frontal (64.3%) or temporal lobe (35.7%) of the left (50%) or right (50%) hemisphere. Pre-operative tractography was reconstructed for each patient, with segmentation of the inferior fronto-occipital fasciculus (IFOF), arcuate fasciculus (AF), uncinate fasciculus (UF), inferior longitudinal fasciculus (ILF), third branch of the superior longitudinal fasciculus (SLF-III), frontal aslant tract (FAT), and cortico-spinal tract (CST). Post-operative percentage of damage and disconnection of each tract, based on the patients' surgical cavities, were correlated with verbal fluencies scores at one week and one month after surgery. Analyses of differences between fluency scores at these timepoints (before surgery, one week and one month after surgery) were performed; lesion-symptom mapping was used to identify the correlation between cortical areas and post-operative scores. Immediately after surgery, a transient impairment of verbal fluency was observed, that improved within a month. Left hemisphere lesions were related to a worse verbal fluency performance, being a damage to the left superior frontal or temporal gyri associated with phonemic or semantic fluency deficit, respectively. At a subcortical level, disconnection analyses revealed that fluency scores were associated to the involvement of the left FAT and the left frontal part of the IFOF for phonemic fluency, and the association was still present one month after surgery. For semantic fluency, the correlation between post-surgery performance emerged for the left AF, UF, ILF and the temporal part of the IFOF, but disappeared at the follow-up. This approach based on the patients' pre-operative tractography, allowed to trace for the first time a dissociation between white matter pathways integrity and verbal fluency after surgery for glioma resection. Our results confirm the involvement of a frontal anterior pathway for phonemic fluency and a ventral temporal pathway for semantic fluency. Finally, our longitudinal results suggest that the frontal executive pathway requires a longer interval to recover compared to the semantic one.
| S-EPMC9400120 | biostudies-literature
Project description:Virtuosity in music performance is often associated with fast, precise, and efficient sound-producing movements. The generation of such highly skilled movements involves complex joint and muscle control by the central nervous system, and depends on the ability to anticipate, segment, and coarticulate motor elements, all within the biomechanical constraints of the human body. When successful, such motor skill should lead to what we characterize as fluency in musical performance. Detecting typical features of fluency could be very useful for technology-enhanced learning systems, assisting and supporting students during their individual practice sessions by giving feedback and helping them to adopt sustainable movement patterns. In this study, we propose to assess fluency in musical performance as the ability to smoothly and efficiently coordinate while accurately performing slow, transitionary, and rapid movements. To this end, the movements of three cello players and three drummers at different levels of skill were recorded with an optical motion capture system, while a wireless electromyography (EMG) system recorded the corresponding muscle activity from relevant landmarks. We analyzed the kinematic and coarticulation characteristics of these recordings separately and then propose a combined model of fluency in musical performance predicting music sophistication. Results suggest that expert performers' movements are characterized by consistently smooth strokes and scaling of muscle phasic coactivation. The explored model of fluency as a function of movement smoothness and coarticulation patterns was shown to be limited by the sample size, but it serves as a proof of concept. Results from this study show the potential of a technology-enhanced objective measure of fluency in musical performance, which could lead to improved practices for aspiring musicians, instructors, and researchers.
| S-EPMC6369163 | biostudies-literature