Perceptual Functions Research Articles

Potential effects of peripheral neuropathy on brain function in patients with type 2 diabetes mellitus

BackgroundThe mechanisms associated between diabetic peripheral neuropathy (DPN) and various brain function abnormalities in patients remains unclear. This study attempted to indirectly evaluate the effect of DPN on brain function in patients with type 2 diabetes mellitus (T2DM) by characterizing the resting-state functional connectivity (FC) of the lower limb sensorimotor cortex (LSM).MethodsForty-four T2DM patients with diabetic peripheral neuropathy (DPN), 39 T2DM patients without diabetic peripheral neuropathy (ND), and 43 healthy controls (HCs) underwent a neuropsychological assessment and resting-state functional magnetic resonance imaging examinations to examine the differences in FC between the LSM and the whole brain. The relationships of FC with clinical/cognitive variables were examined.ResultsIn comparison with the HCs group, the ND group showed reduced FC of the LSM with the right lateral occipitotemporal cortex (LOTC) and increased FC with the medial superior frontal gyrus (SFGmed), while the DPN group showed reduced FC of the LSM with the right cerebellar lobule VI, the right LOTC, the rostral prefrontal cortex (rPFC), and the anterior cingulate gyrus (ACC). Moreover, in comparison with the ND group, the DPN group showed reduced FC of the LSM with the ACC, SFGmed, and rPFC. In the DPN group, the FC between the LSM and right cerebellar lobule VI was significantly correlated with fasting blood glucose levels (r = -0.490, p = 0.001), and that between the LSM and ACC was significantly correlated with the Montreal Cognitive Assessment score (r = 0.479, p = 0.001).ConclusionsPatients with T2DM may show abnormal motion-related visual perceptual function before the appearance of DPN. Importantly, DPN can influence the brain regions that maintain motion and motor control, and this effect is not limited to motor function, which may be the central neuropathological basis for diabetic peripheral neuropathy.

Enhancing Image Classification with Super-Resolution Techniques: A Comparative Study of SRResNet and SRGAN

Abstract. This paper explores the use of super-resolution (SR) approaches to enhance the performance of image classification, with particular attention to the effects of various SR models on classification accuracy. Using the ImageNet Dogs dataset for assessment, the paper examines two SR techniques: Super Resolution Generative Adversarial Networks (SRGAN) and Super Resolution Residual Networks (SRResNet). The research demonstrates that deep learning-based SR methods can enhance classification accuracy. Notably, SRResNet is identified as the superior method for improving classification performance, despite generating less visually appealing images compared to SRGAN. These finding highlights that while Generative Adversarial Networks (GANs) and perceptual loss functions can enhance image quality, their impact on classification accuracy may not always be substantial. The study offers important new perspectives on the relative merits of different SR approaches, highlighting the necessity of choosing the right SR methods in accordance with the particular demands of picture classification tasks. The results suggest that SRResNet, with its focus on accuracy over visual appeal, is more effective for boosting classification model performance, offering guidance for optimizing SR methods in practical image classification applications.

A systematic review of deep learning-based denoising for low-dose computed tomography from a perceptual quality perspective.

Low-dose computed tomography (LDCT) scans are essential in reducing radiation exposure but often suffer from significant image noise that can impair diagnostic accuracy. While deep learning approaches have enhanced LDCT denoising capabilities, the predominant reliance on objective metrics like PSNR and SSIM has resulted in over-smoothed images that lack critical detail. This paper explores advanced deep learning methods tailored specifically to improve perceptual quality in LDCT images, focusing on generating diagnostic-quality images preferred in clinical practice. We review and compare current methodologies, including perceptual loss functions and generative adversarial networks, addressing the significant limitations of current benchmarks and the subjective nature of perceptual quality evaluation. Through a systematic analysis, this study underscores the urgent need for developing methods that balance both perceptual and diagnostic quality, proposing new directions for future research in the field.

Thematic and intertextual analysis from a feasibility study of the Bonny Method of Guided Imagery and Music with clients in eating disorder treatment

IntroductionEating disorders (ED) are characterized by serious and persistent disturbances with eating, weightcontrol, and body image. Symptoms impact physical health, psychosocial functioning, and can be life-threatening. Individuals diagnosed with an ED experience numerous medical and psychiatric comorbidities due to issues caused by or underlying the ED. Therefore, it is vital to address the complex nature of an ED, as well as the comorbid and underlying issues. This necessitates a psychotherapeutic approach that can help to uncover, explore, and support working through unresolved emotions and experiences. Guided Imagery and Music (GIM) is an in-depth music psychotherapy approach utilizing therapist-programmed music to support the client in uncovering and examining underlying and unresolved issues. The literature surrounding the use of GIM with clients in ED treatment is anecdotal and comprised primarily of clinical case studies.MethodThis secondary analysis, based on a descriptive feasibility study that integrated GIM sessions into the client’s regular ED treatment and examined 116 transcripts from a series of sessions of eight clients.ResultsThematic analysis of the transcripts identified nine subthemes and three themes that emerged. These themes include emotional landscape (feeling stuck, acknowledging emotions, and working through unresolved emotions), relationships (self, others, and eating disorders), and transformation and growth (finding strength, change, and empowerment). A short series of GIM sessions helped ED clients identify and address issues underlying the ED and to gain or reclaim a sense of self that enabled them to make choices for their life that support their recovery and sense of empowerment. Intertextual analysis revealed imagery indicative of the Hero’s Journey.DiscussionFurther, how engagement in this embodied aesthetic experience stimulates perceptual, cognitive, and affective brain functions which are key in fostering behavioural and psychological change is explicated as it relates to ED treatment and recovery.

Towards Human Sensory Augmentation: A Cognitive Neuroscience Framework for Evaluating Integration of New Signals within Perception, Brain Representations, and Subjective Experience

New wearable devices and technologies provide unprecedented scope to augment or substitute human perceptual abilities. However, the flexibility to reorganize brain processing to use novel sensory signals during early sensitive periods in infancy is much less evident at later ages, making integration of new signals into adults’ perception a significant challenge. We believe that an approach informed by cognitive neuroscience is crucial for maximizing the true potential of new sensory technologies. Here, we present a framework for measuring and evaluating the extent to which new signals are integrated within existing structures of perception and experience. As our testbed, we use laboratory tasks in which healthy volunteers learn new, augmented perceptual-motor skills. We describe a suite of measures of (i) perceptual function (psychophysics), (ii) neural representations (fMRI/decoding), and (iii) subjective experience (qualitative interview/micro-phenomenology) targeted at testing hypotheses about how newly learned signals become integrated within perception and experience. As proof of concept, we provide example data showing how this approach allows us to measure changes in perception, neural processing, and subjective experience. We argue that this framework, in concert with targeted approaches to optimizing training and learning, provides the tools needed to develop and optimize new approaches to human sensory augmentation and substitution.

DuINet: A dual-branch network with information exchange and perceptual loss for enhanced image denoising

Image denoising is a fundamental task in image processing and low-level computer vision, often necessitating a delicate balance between noise removal and the preservation of fine details. In recent years, deep learning approaches, particularly those utilizing various neural network architectures, have shown significant promise in addressing this challenge. In this study, we propose DuINet, a novel dual-branch network specifically designed to capture complementary aspects of image information. DuINet integrates an information exchange module that facilitates effective feature sharing between the branches, and it incorporates a perceptual loss function aimed at enhancing the visual quality of the denoised images. Extensive experimental results demonstrate that DuINet surpasses existing dual-branch models and several state-of-the-art convolutional neural network (CNN)-based methods, particularly under conditions of severe noise where preserving fine details and textures is critical. Moreover, DuINet maintains competitive performance in terms of the LPIPS index when compared to deeper or larger networks such as Restormer and MIRNet, underscoring its ability to deliver high visual quality in denoised images.

Blockchain for video watermarking: An enhanced copyright protection approach for video forensics based on perceptual hash function.

As a direct result of advancements in digital technology and the Internet, the copyright protection and information integrity of multimedia that are being published across the Internet have emerged as a major and urgent issue that needs to be addressed. The technique of digital watermarking may be used to protect intellectual property. In terms of authentication, resilience, storage, and capacity of digital watermarking information, there is still room for development. Blockchain's potential in video copyright protection and management applications has motivated researchers. Copyright owners and consumers may now communicate directly via copyright protection apps built on the blockchain, eliminating the need for distributers and the associated fees. Nonetheless, the current blockchain-based video watermarking solutions require storing a significant number of coordinates depending on the watermark size and are susceptible to video frame attacks on the video frame texture region. This study proposes an enhanced video copyright management approach that incorporates digital watermarking, the blockchain, and a perceptual hash function. Watermark information is stored on a blockchain structure, which also acts as a timestamp for verification purposes. To verify watermark data without the source video, a perceptual hash function is employed to compute a hash value based on the structural information of video frames. The contribution is in learning how to extract a hash function from a small number of video frames that still adequately represent a large amount of video content while also reducing the number of unnecessary video frames and the amount of computation required to summarize and index that content in a blockchain. This expedites the dissemination of copyrighted works and increases their security and readability, hence facilitating their circulation on the Internet. Our experimental results demonstrate that this approach is memory efficient, as it only needs to store one key for each key frame, regardless of the size of the watermark. Additionally, the overall robustness is greatly improved by using the blockchain's random hash function. Therefore, new and important advancements in video watermarking have been realized because of this effort.

LLD-GAN: An end-to-end network for low-light image demosaicking

Demosaicking of low and ultra-low light images has wide applications in the fields of consumer electronics, security, and industrial machine vision. Denoising is a challenge in the demosaicking process. This study introduces a comprehensive end-to-end low-light demosaicking framework called LLD-GAN (Low Light Demosaicking Generative Adversarial Network), which greatly reduces the computational complexity. Our architecture employs a Wasserstein GAN framework enhanced by a gradient penalty mechanism. We have redesigned the generator based on the UNet++ network as well as its corresponding discriminator, which makes the model learning more efficient. In addition, we propose a new loss metric grounded in the principles of perceptual loss to obtain images with better visual quality. The contribution of Wasserstein GAN with gradient penalty and perceptual loss function was proved to be beneficial by our ablation experiments. For RGB images, we tested the proposed model under a wide range of low light levels, from 1/30 to 1/150 of normal light level, for 16-bit images with added noise. For actual low-light raw sensor images, the model was evaluated under three distinct lighting conditions: 1/100, 1/250, and 1/300 of normal exposure. The qualitative and quantitative comparison against advanced techniques demonstrates the validity and superiority of the LLD-GAN as a unified denoising-demosaicking tool.

Diversity in perceptual, social, and executive functions in preschoolers from Germany and Iran

The cultural background and age shape the cognitive and social development of children. This study aimed to compare perceptual, executive, and social functioning in two age groups of preschoolers from Germany and Iran. One hundred sixty-five children (83 Iranian and 82 German) participated in the study, with 87 children aged three years and 78 children aged four years. The participants completed a series of tasks to assess different cognitive functions, including the theory of mind as a measure of social cognition, picture mental rotation test to evaluate spatial ability, and several tests including digit span, Corsi block tapping, day-night, grass-snow, dimensional change card sort, and active and passive vocabulary tests to assess executive functions. The findings revealed that German children performed better than their Iranian peers regarding spatial ability and theory of mind. Additionally, German children outperformed Iranian children in working memory, cognitive flexibility, and verbal fluency, while Iranian children demonstrated better inhibitory control. Furthermore, the correlational analysis indicated that in German children, executive function correlated with the theory of mind, whereas in Iranian children, executive function was associated with perceptual functions. Age impacted the results. These results highlight the role of cultural factors in shaping cognitive functioning and emphasize the need to consider cultural influences when examining cognitive development in different populations.

Amnesia and homework

Introduction: In humans, the phenomenon of temporally graded retrograde amnesia has been described in the clinic and the laboratory for more than 100 years. The memory f\deficit can occur against a background of intact intellectual and perceptual functions. Material and methods: There were analyzed articles from the PubMed database from the last 5 years 2019-2024, mentioning such words as “amnesia”, “homework”. Results: Amnestic syndrome is narrower than once believed in the sense that a number of learning and memory abilities are preserved. The brain system damage in amnesia has only a temporary role in memory. In the 1950 edition of the Encyclopedia of Educational Research H.J. Otto wrote “compulsory homework does not result in sufficiently improved academic accomplishments to justify retention” [1]. Eighteen years later P.R Wildman went a step further, starting that “whenever homework crowds out social experience, outdoor recreation, and creative activities and whenever it usurps time devoted to sleep, it is not meeting the basic needs of children and adolescents” [2]. Conclusion: Thus, as Otto’s remark exemplifies, homework fell out of favor. Homework, it was believed could accelerate knowledge acquisition. Wildman expressed the concern of the time that too much emphasis on school would lead to the neglect of other areas of personal fulfillment. Not all amnesias, results from brain damage. Amnesia may be produced by severe emotional trauma or naturally occurring changes in state or arousal [3-5]. Improves creative activity total quality management as a predictor of artificial intelligence provides competitive marketing, proven and by the role of social networks in reducing stress [6,7], confirming that amnesia has only a temporary role in memory.

Robust Metallized Liquid Crystal Elastomer Fiber Arrays Toward a Machine Learning‐Assisted Artificial Neuromuscular System with Perceptual Function

AbstractEndowing artificial muscles with perceptual function, as an octopus does, is highly desired but still suffering from interfacing mismatch between actuating and sensing units in a thin fiber. Herein, an artificial neuromuscular fiber capable of electrically responding to external strain/temperature and actuation path with power supply is reported by using polymer‐assisted metal deposition to firmly coat Cu nanoparticles on the surface of liquid crystal elastomer (LCE). The LCE core acts as an actuator, while the wrinkled Cu sheath provides environment interaction and actuation monitoring. Benefiting from the levodopa/polyethyleneimine interface design, this fiber exhibits large reversible contraction (47.61%), fast strain rate (370% s−1), high output power density (663.75 W kg−1) and reliable durability (1000 cycles) under electrical stimulation. Furthermore, by combining as‐made fiber arrays with electronic system and computing algorithm, this machine learning‐assisted artificial neuromuscular system can actuate the Chinese shadow puppetry and recognize its body movements with high accuracy. This work paves a revolutionary way for fabricating next‐generation flexible robots.

Effects of Combined Visual-Motor Response Training on Cognitive Function and Brain Plasticity Mechanisms in Various Populations: Protocol for a Single-Center, Open-Label, Controlled Clinical Trial.

Cognitive impairment is one of the major diseases facing the aging population. The progressive decline of cognitive function can lead to declining health or even the loss of life, work, and social ability. Exercise and behavioral stimulation can increase neurotransmitters in the brain and improve overall health and cognitive function. Reactivity training can mobilize neuromuscular function and induce changes in brain plasticity, which may effectively improve cognitive dysfunction and delay the occurrence and development of Alzheimer disease; however, the evidence supporting its effectiveness is still limited. This study aims to explore the effectiveness and reliability of visual-motor reaction training in improving cognitive function, thereby promoting the application of novel nonpharmacological therapies. This study is a single-center, open-label, controlled clinical trial. A total of 78 participants will be recruited for the study, including an equal number of athletes, ordinary healthy college students, and ordinary older adults in the community. Participants will receive 2 weeks of visual-motor response training. The primary outcome of this study is to assess differences in functional magnetic resonance imaging (fMRI) at 2 weeks. The secondary outcomes were the following: acousto-optic response time, Hamilton Depression Rating Scale (HAM-D), Hamilton Anxiety Rating Scale (HAM-A), Mini Mental State Examination (MMSE), Activity of Daily Living (ADL) Scale, Subjective Cognitive Decline Questionnaire-9 (SCD-Q9), a 10-word memory test, and safety. The study was approved by the Shanghai Clinical Research Ethics Committee on January 2, 2024 (SECCR/2023-162-01). As of September 11, 2024, we have completed the recruitment of all 3 groups of volunteers. We expect to complete data collection and analysis by February 2025. The purpose of this study is to compare improvements in brain perceptual motor functions and cognitive levels across different populations through response ability training and to explore the efficacy and safety of exercise-based nonpharmacological therapies in improving cognitive function. Other potential benefits include understanding the functional differences and perceptual characteristics of the brain's perceptual-motor system between athletes and the general population and exploring the adaptability of the brain in acquiring skills during competitive sports training. This could provide an evidence base for early sports talent development and broader youth development. Chinese Clinical Trial Registry ChiCTR2400079602; https://tinyurl.com/23fbbndw. DERR1-10.2196/56424.

All-In-One Optoelectronic Transistors for Bio-Inspired Visual System.

Visual perception has profound effects on human decision-making and emotional responses. Replicating the functions of the human visual system through device development has been a constant pursuit in recent years. However, to fully simulate the various functions of the human visual system, it is often necessary to integrate multiple devices with different functions, resulting in complex, large-volume device structures and increased power consumption. Here, an optoelectronic transistor with comprehensive visual functions is introduced. By coupling diverse photoreceptive properties of the channel and electrical regulation through charge injection/ferroelectric switching from the hafnium-based gate, the devices can simulate functions of both photoreceptors in the retina and synapses in the visual cortex. A device array is constructed to confirm the perceptual functions of cone and rod cells. Subsequently, color discrimination and recognition for color images are achieved by combining the tunable perception and synapse functions. Then an intelligent traffic judgment system with this all-in-one device is developed, which is capable of making judgments and decisions regarding traffic signals and pedestrian movements. This work provides a potential solution for developing compact and efficient devices for the next-generation bio-inspired visual system.

Switching to Low Neurotoxic Antiretrovirals to Improve Neurocognition Among People Living With HIV-1-Associated Neurocognitive Disorder: The MARAND-X Randomized Clinical Trial

Background: The pathogenesis of HIV-associated neurocognitive (NC) impairment is multifactorial, and antiretroviral (ARV) neurotoxicity may contribute. However, interventional pharmacological studies are limited. Methods: Single-blind, randomized (1:1), controlled trial to assess the change of NC performance (Global Deficit Score, GDS, and domain scores) in PLWH with NC impairment randomized to continue their standard of care treatment or to switch to a less neurotoxic ARV regimen: darunavir/cobicistat, maraviroc, emtricitabine (MARAND-X). Participants had plasma and cerebrospinal fluid HIV RNA< 50 copies/mL, R5-tropic HIV, and were on ARV regimens that did not include efavirenz and darunavir. The change of resting-state electroencephalography was also evaluated. The outcomes were assessed at week 24 of the intervention through tests for longitudinal paired data and mixed-effect models. Results: Thirty-eight participants were enrolled and 28 completed the follow-up. Global Deficit Score improved over time but with no difference between arms in longitudinal adjusted models. Perceptual functions improved in the MARAND-X, while long-term memory improved only in participants within the MARAND-X for whom the central nervous system penetration-effectiveness (CNS penetration effectiveness) score increased by ≥3. No significant changes in resting-state electroencephalography were observed. Conclusions: In this small but well-controlled study, the use of less neurotoxic ARV showed no major beneficial effect over an unchanged regimen. The beneficial effects on the memory domain of increasing CNS penetration effectiveness score suggest that ARV neuropenetration may have a role in cognitive function.

Enhancement of perceptual and cognitive functions in near-death experience: A perspective from embodiment theories

The phenomenon of near-death experience (NDE) is attracting a growing attention among researchers of various fields. In this study, we looked at NDE from a cognitive perspective to find out how NDE events are embodied when people recall and describe them. We examined the descriptions of a group of people talking about what they had experienced in the state of NDE. Based on the gesture-as-simulated-action theory, we assumed that co-speech gestures occurring when people were talking about their experiences were physical realizations of mentally-simulated events. The results showed that the number of iconic and metaphoric gestures occurring with expressions referring to NDE events was significantly larger than those occurred with expressions referring to ordinary events. Based on these results, we suggest that embodied memory for NDE events is stronger than embodied memory for ordinary non-NDE events. NDE events are perceived, recalled, and embodied by strong activation of sensorimotor systems in a state of perceptual and cognitive enhancement, although sensorimotor systems seem to be unresponsive to sensory stimuli during NDE. Finally, based on theories of embodied cognition, we conclude that the enhancement of cognitive functions that takes place in the state of NDE is the result of enhancement in perceptual functions.

Overview of High-Dynamic-Range Image Quality Assessment.

In recent years, the High-Dynamic-Range (HDR) image has gained widespread popularity across various domains, such as the security, multimedia, and biomedical fields, owing to its ability to deliver an authentic visual experience. However, the extensive dynamic range and rich detail in HDR images present challenges in assessing their quality. Therefore, current efforts involve constructing subjective databases and proposing objective quality assessment metrics to achieve an efficient HDR Image Quality Assessment (IQA). Recognizing the absence of a systematic overview of these approaches, this paper provides a comprehensive survey of both subjective and objective HDR IQA methods. Specifically, we review 7 subjective HDR IQA databases and 12 objective HDR IQA metrics. In addition, we conduct a statistical analysis of 9 IQA algorithms, incorporating 3 perceptual mapping functions. Our findings highlight two main areas for improvement. Firstly, the size and diversity of HDR IQA subjective databases should be significantly increased, encompassing a broader range of distortion types. Secondly, objective quality assessment algorithms need to identify more generalizable perceptual mapping approaches and feature extraction methods to enhance their robustness and applicability. Furthermore, this paper aims to serve as a valuable resource for researchers by discussing the limitations of current methodologies and potential research directions in the future.

Environmental Lighting Conditions, Phenomenal Contrast, and the Conscious Perception of Near and Far.

Recent evidence in systems neuroscience suggests that lighting conditions affect the whole chain of brain processing, from retina to high-level cortical networks, for perceptual and cognitive function. Here, visual adaptation levels to three different environmental lighting conditions, (1) darkness, (2) daylight, and (3) prolonged exposure to very bright light akin to sunlight, were simulated in lab to investigate the effects of light adaptation levels on classic cases of subjective contrast, assimilation, and contrast-induced relative depth in achromatic, i.e., ON-OFF pathway mediated visual configurations. After adaptation/exposure to a given lighting condition, configurations were shown in grouped and ungrouped conditions in random order to healthy young humans in computer-controlled two-alternative forced-choice procedures that consisted of deciding, as quickly as possible, which of two background patterns in a given configuration of achromatic contrast appeared lighter, or which of two foreground patterns appeared to stand out in front, as if it were nearer to the observer. We found a statistically significant effect of the adaptation levels on the consciously perceived subjective contrast (F(2,23) = 20.73; p < 0.001) and the relative depth (F(2,23) = 12.67; p < 0.001), a statistically significant interaction between the adaptation levels and the grouping factor (F(2,23) = 4.73; p < 0.05) on subjective contrast, and a statistically significant effect of the grouping factor on the relative depth (F(2,23) = 13.71; p < 0.01). Visual adaption to different lighting conditions significantly alters the conscious perception of contrast and assimilation, classically linked to non-linear functional synergies between ON and OFF processing channels in the visual brain, and modulates the repeatedly demonstrated effectiveness of luminance contrast as a depth cue; the physically brighter pattern regions in the configurations are no longer consistently perceived as nearer to a conscious observer under daylight and extreme bright light adapted (rod-saturated) conditions.

Visual Performance of People With Albinism Assessed With Generalizable and Adaptive AIM and FInD Methods.

People with albinism (PwA) are known to have visual impairments; however, little is known about whether these functions are disrupted across earlier and later stages of the visual pathway. We investigated distinct perceptual functions and fixation stability within each observer and compared the data with age- (±5years) and sex-matched controls. Twenty-one self-reported PwA and twenty-one controls were recruited. Angular-indication measurement (AIM) and foraging-interactive-D-prime (FInD) psychophysical methods were deployed to measure OS, OD, and OU near visual acuity, spatial contrast sensitivity function (CSF), temporal contrast sensitivity (tCS; 0.5 c/°; horizontal grating: 0, 1, 2, 4, and 8Hz), OU glare acuity, threshold-versus-contrast (2c/° vertical grating), long, medium, and short wavelength cone-isolated color detection, color discrimination, stereoacuity across spatial frequencies (1c/°, 2c/°, 4c/°, 8c/°), horizontal, circular, radial pattern and motion coherence, and equivalent-noise motion detection. Thresholds were determined by AIM and FInD and compared using N-ANOVAs, t-tests, planned multi-comparisons, correlations, and unsupervised, agglomerative hierarchical cluster analysis for each group. We found significant differences between groups for most visual functions except for simple and complex form-coherence (two way-ANOVAs, P > 0.05) and complex motion coherence. Correlations between outcomes revealed more significant correlations for PwA and differences in the specific correlates between groups. Unsupervised hierarchical clustering revealed different functional clusters between groups. AIM and FInD successfully interrogated visual deficits in PwA. Overall, PwA showed impaired performance in achromatic, chromatic, temporal, and binocular functions, and had higher intrinsic noise levels. Midlevel vision was comparable between groups. Unsupervised cluster analysis and correlation between outcomes revealed a difference in functional outcome clusters between groups. The results may help to increase the efficiency of screening and identify target deficits for rehabilitation.

GLFuse: A Global and Local Four-Branch Feature Extraction Network for Infrared and Visible Image Fusion

Infrared and visible image fusion integrates complementary information from different modalities into a single image, providing sufficient imaging information for scene interpretation and downstream target recognition tasks. However, existing fusion methods often focus only on highlighting salient targets or preserving scene details, failing to effectively combine entire features from different modalities during the fusion process, resulting in underutilized features and poor overall fusion effects. To address these challenges, a global and local four-branch feature extraction image fusion network (GLFuse) is proposed. On one hand, the Super Token Transformer (STT) block, which is capable of rapidly sampling and predicting super tokens, is utilized to capture global features in the scene. On the other hand, a Detail Extraction Block (DEB) is developed to extract local features in the scene. Additionally, two feature fusion modules, namely the Attention-based Feature Selection Fusion Module (ASFM) and the Dual Attention Fusion Module (DAFM), are designed to facilitate selective fusion of features from different modalities. Of more importance, the various perceptual information of feature maps learned from different modality images at the different layers of a network is investigated to design a perceptual loss function to better restore scene detail information and highlight salient targets by treating the perceptual information separately. Extensive experiments confirm that GLFuse exhibits excellent performance in both subjective and objective evaluations. It deserves note that GLFuse effectively improves downstream target detection performance on a unified benchmark.

Mapping the structure-function relationship along macroscale gradients in the human brain

Functional coactivation between human brain regions is partly explained by white matter connections; however, how the structure-function relationship varies by function remains unclear. Here, we reference large data repositories to compute maps of structure-function correspondence across hundreds of specific functions and brain regions. We use natural language processing to accurately predict structure-function correspondence for specific functions and to identify macroscale gradients across the brain that correlate with structure-function correspondence as well as cortical thickness. Our findings suggest structure-function correspondence unfolds along a sensory-fugal organizational axis, with higher correspondence in primary sensory and motor cortex for perceptual and motor functions, and lower correspondence in association cortex for cognitive functions. Our study bridges neuroscience and natural language to describe how structure-function coupling varies by region and function in the brain, offering insight into the diversity and evolution of neural network properties.