Fine-scale spatial dynamics within functional brain networks manifest as high spatial-frequency variations that conventional independent component analysis (ICA) methods fail to capture. These subtle changes may carry critical information about transient connectivity and disordered brain function. We developed NeuroMark-DyFICA (dynamic frequency-informed ICA), a novel framework to enhance detection of spatiotemporal variability in fMRI data. It integrates three stages: dynamic NeuroMark ICA across sliding windows to estimate time-varying, spatially constrained networks; high-pass spatial filtering to emphasize fine-scale spatial features; and group-level ICA to extract refined dynamic components with subject-specific mixing weights. Unlike prior NeuroMark applications or conventional dynamic ICA, NeuroMark-DyFICA establishes a reproducible latent space of high-frequency dynamics, uniquely capturing transient, fine-scale reconfigurations of network topography. Validation using a controlled 2D simulation demonstrated reliable detection of subtle spatial shifts mimicking pathology, which conventional ICA failed to recover. Applying to resting-state fMRI from schizophrenia patients and healthy controls, multiple networks were estimated. We highlight six representative systems (thalamus, auditory, visual/fusiform, middle frontal, default mode, cerebellum). Results revealed two complementary abnormalities in schizophrenia: an imbalance between inactive and hyper-engaged states and altered convergence among dynamic states. NeuroMark-DyFICA reveals fine-grained spatiotemporal disruptions in brain networks, offering mechanistic insights and potential biomarkers for psychiatric disorders. Brain network dynamics are most often studied as changes in the expression or interaction of large-scale, canonical patterns over time. While this perspective has been highly productive, it implicitly emphasizes low spatial-frequency variations and may overlook more focal, transient reconfigurations that occur within otherwise stable networks. Neurophysiological and clinical evidence increasingly suggests that cognition- and disease-related alterations can manifest as subtle, localized shifts in network boundaries, internal structure, or spatial coherence—features that reside in the high spatial-frequency domain and are difficult to capture with conventional dynamic approaches. In this study, we introduce NeuroMark-DyFICA (dynamic frequency-informed independent component analysis), a framework designed to capture dynamic, fine-scale spatial variations in functional brain networks while preserving biologically meaningful network structure. The method integrates template-guided independent component analysis with spatial-frequency filtering and group-level decomposition, enabling isolation of transient, high-frequency spatial changes within canonical networks in a fully automated and reproducible manner. Using controlled simulations, we demonstrate that NeuroMark-DyFICA is sensitive to small spatial shifts that conventional static or low-frequency–focused methods often overlook. When applied to resting-state fMRI data from individuals with schizophrenia and healthy controls, the framework reveals abnormal dynamic spatial patterns across auditory, visual, frontal, default-mode, and cerebellar networks, including altered prevalence and coordination of dynamic spatial states. By explicitly targeting fine-scale spatial dynamics, NeuroMark-DyFICA provides a new lens for studying how brain networks reconfigure over time and how these processes are disrupted in neuropsychiatric disorders. This approach offers a principled pathway toward more sensitive biomarkers and advances precision neuroscience by bridging large-scale network organization with localized, clinically relevant variability.
Sleep-disordered breathing (SDB) is common in childhood and is associated with attentional and behavioral impairments despite largely preserved sleep macrostructure and minimal abnormalities in conventional electroencephalographic measures. This discrepancy has contributed to the perception that sleep is relatively preserved in pediatric SDB and has limited understanding of the physiological mechanisms underlying morbidity. To determine whether pediatric SDB is associated with disruption of the regional organization and homeostatic dynamics of slow-wave activity (SWA), a key physiological marker of sleep- dependent neural recovery and development. Cross-sectional study of 62 children aged 4 to 12 years who underwent overnight polysomnography with high-density electroencephalography in a laboratory setting. Participants were recruited from clinical referrals and the community, spanning the full spectrum of SDB severity. SDB severity indexed by hypopnea index (HI), apnea-hypopnea index (AHI), and obstructive apnea index (OAI). Regional electroencephalogram-derived SWA (0.5-4 Hz) topography and exponential decay parameters derived from frontal and posterior cortical regions. The frontal-to-posterior decay- rate ratio was evaluated as a summary measure of regional sleep homeostasis. In children with lower hypopnea index, SWA demonstrated the expected developmental pattern, with posterior predominance in younger children and a progressive shift toward a more balanced anterior-posterior distribution with age. Increasing HI was associated with attenuation or reversal of this spatial organization. Global SWA showed no meaningful association with SDB severity. In contrast, regional frontal and posterior decay parameters were strongly associated with HI (adjusted R² = 0.53; p < 1 × 10⁻⁶) but not OAI (adjusted R² = 0.05; p = .95). The frontal- to-posterior decay-rate ratio showed the strongest association with HI β = 4.15; 95% CI, 3.17- 5.13; p < 1 × 10⁻¹⁰; adjusted R² = 0.55. Pediatric SDB was associated with regional disruption of slow-wave sleep homeostasis rather than global loss of deep sleep. These alterations affected both the spatial organization and temporal dynamics of SWA during a period of active cortical maturation and were not captured by conventional sleep metrics. Regional SWA dynamics may provide a developmentally sensitive marker of physiological disease burden in children with SDB. Question: Does pediatric sleep-disordered breathing disrupt the regional organization and homeostatic dynamics of slow-wave activity during development in ways that are not captured by conventional sleep metrics?Findings: In this cross-sectional study of 62 children across the spectrum of sleep-disordered breathing, hypopnea burden was associated with altered regional organization and overnight dissipation of NREM slow-wave activity (SWA) despite preserved global SWA. A frontal-to- posterior SWA decay-rate ratio was strongly associated with hypopnea severity, whereas global SWA was not.Meaning: Pediatric sleep-disordered breathing may disrupt sleep physiology in a regional, developmentally meaningful manner not captured by conventional polysomnography, suggesting a potential physiological marker of disease burden beyond event counts.
Functional connectivity is often constructed to understand functional organizations in neural systems, where the brain regions and their pairwise interactions are viewed as nodes and edges, respectively. In practice, functional connectivity is commonly estimated via the correlation of pairs of brain regions. One limitation is that the correlation coefficient captures only the pairwise linear dependence relationship between pairs of nodes and may fail to capture complex higher-order relationships. Recently, a novel concept known as edge-centric functional connectivity (eFC) has been introduced to measure interactions between pairs of edges based on the cofluctuation of two nodal time series, offering a new perspective for understanding brain networks. Nevertheless, eFC considers the absolute levels of edge time series, that is, their mean values, in estimation. If the parameter of interest is the covariation between a pair of edges, incorporating mean values of edge time series can introduce bias or deviation, resulting in a skewed estimation. In this manuscript, we propose an alternative approach to estimate the unbiased covariation between pairs of edges, termed centered edge functional connectivity (ceFC), with theoretical foundations. We demonstrate that the proposed estimator is consistent with a sufficient sample size or number of time frames. Additionally, we develop a multiple hypothesis testing framework with a controlled false discovery rate to evaluate the strength of the unbiased covariation among edges. Furthermore, we employ thresholding to obtain a thresholded estimator that has been shown to converge to the true ceFC matrix in high-dimensional settings in which the number of nodes is much larger than the number of samples or time frames. We validate the finite sample performance of the proposed methods via numerical studies and a data application using the Midnight Scan Club dataset. Functional connectivity typically examines the correlations between pairs of brain nodes, which only capture pairwise linear dependence. Recently, the edge-centric functional connectivity (eFC) method has shifted the focus toward the relationships between pairs of edges. While promising, this current method can introduce statistical biases that skew our understanding of these higher-order interactions. In this study, we propose centered edge functional connectivity (ceFC), a framework designed to provide an unbiased view of how brain edges covary. By establishing a rigorous statistical foundation and an inference framework to identify significant interactions, our method ensures reliable results even with a limited sample size. We show that our approach can help identify more stable overlapping brain community structures, providing a new perspective on brain organization.
Understanding how the brain gives rise to social cognition has been a key goal of neuroimaging research. Both changes in regional activation as well as functional connectivity have been implicated as potential mechanisms underlying social cognition, but the two have rarely been examined concurrently. Moreover, because the neural processes underlying social cognition are dynamic, developing approaches to capture dynamic changes in regional activity and functional connectivity are critical. Here, we describe a novel analysis approach that captures both regional activity and dynamic functional connectivity simultaneously during a naturalistic, socially focused movie-watching task. We found that both regional activation and functional connectivity were uniquely related to awkwardness, a judgment associated with social faux pas detection and theory of mind. Regional activation within sensorimotor networks was positively associated with awkwardness, whereas activation in the default network was negatively associated. Models including functional connectivity accounted for unique variance beyond models with activity alone. Specifically, dynamic functional connectivity between networks, primarily the frontoparietal control network, was positively associated with awkwardness. Together, these findings suggest that both dynamic regional brain activity and functional connectivity each uniquely contribute to complex and dynamic social judgments. We assessed the relationship between regional activity, functional connectivity, and awkwardness judgments during a video-watching task using fMRI in a cohort of healthy young adults. We used a novel analysis framework to simultaneously disentangle the unique contributions of dynamic regional activity and dynamic functional connectivity in their predictions of awkwardness. We found regional activity associated with awkwardness primarily in sensorimotor, frontoparietal control, and default networks, and functional connectivity associated with awkwardness within and between the visual, dorsal attention, frontoparietal control, and default networks. Finally, hierarchical models demonstrated that modeling functional connectivity significantly explained additional variance beyond modeling the activity of two regions alone, indicating the unique explanatory power offered by dynamic functional connectivity in relation to behavior.
In robotics research, each subdomain presents a distinct set of challenges, and any framework designed for a given domain must effectively address these complexities. However, a single application within that domain may not fully capture the breadth of challenges inherent to it. To enable systematic and comprehensive evaluation, the robotics community has developed standardized problem scenarios and associated performance metrics, commonly referred to as benchmarks, which collectively represent the diverse challenges arising across applications. In this work, we evaluate our task-motion Planning (TMP) framework on five benchmarks proposed by the TMP community. We begin by briefly describing our iterative deepening AND/OR graph-based TMP planner. Subsequently, we assess its performance across these benchmarks, each designed to capture different aspects of the challenges in TMP. The evaluation demonstrates that the proposed planner successfully solves all five benchmarks, thereby indicating that our framework constitutes a robust and effective solution for TMP.
Propane/propylene splitting is among the most energy- and capital-intensive separations in the petrochemical industry, yet it is indispensable for upgrading mixed C3 cuts to polymer-grade propylene. Because propane and propylene exhibit nearly overlapping physicochemical properties, conventional cryogenic distillation requires very large columns, high reflux ratios, and substantial thermal duties. Adsorption-based separations using porous solids offer a compelling non-cryogenic alternative, especially when the adsorbent is propane-selective (inverse-selective). Propane is retained as the impurity and high-purity propylene can be delivered directly as the raffinate, potentially simplifying process flowsheets and reducing regeneration penalties associated with product-capture schemes. This Review summarizes recent advances, primarily from 2019 to early 2026, in propane-selective porous materials for inverse propane/propylene separation, with metal-organic frameworks (MOFs) as the central focus and porous carbons, hydrogen-bonded organic frameworks (HOFs), and covalent organic frameworks (COFs) discussed as complementary families, while earlier landmark studies such as ZIF-8 are included where necessary for historical context. We highlight how pore-surface polarity/hydrophobicity, ultramicropore confinement, cooperative weak interactions, and framework dynamics can invert the conventional olefin preference and enhance propane capture within practical operating windows. Emphasis is placed on application-relevant evaluation, including gas adsorption, fixed-bed breakthrough behavior, working capacity, regenerability, tolerance to humidity and trace impurities, cycling durability, and prospects for scalable shaping and process integration.
Aerosol liquid water content (ALWC) plays an important role in climate and public health by influencing aerosol formation, chemical composition, and toxicity. However, ALWC remains sparsely measured and poorly constrained across space and time, despite its large variability. In this study, we derived a high-resolution (1 km × 1 km, daily) ALWC dataset for the contiguous US from 2000 to 2019. The dataset was generated by training machine learning (ML) models on outputs from a chemical transport model (GEOS-Chem) to capture the thermodynamic relationships between ALWC and relevant predictors, then applying these relationships to high-resolution, biased-corrected input datasets. Compared with GEOS-Chem simulations, the ML-based dataset better captures daily variations and spatial heterogeneity in ALWC. The predicted ALWC levels are highest in the Midwest US and lowest in the Western US, largely driven by regional differences in PM2.5 concentration, chemical composition, temperature, and relative humidity. Over the study period, ALWC declined significantly across most regions, driven primarily by the reduction in sulfate. We further demonstrate that ALWC provides a physically meaningful constraint for interpreting variability in water-soluble iron, a health-relevant fraction of aerosol metals, highlighting the potential value of this dataset for future studies of aerosol toxicity and epidemiological exposure.
Accurate quantification of local strain fields during bladder contraction is essential for understanding the biomechanics of bladder micturition, in both health and disease. Conventional digital image correlation (DIC) methods have been successfully applied to various biological tissues; however, this approach requires artificial speckling, which can alter both passive and active properties of the tissue. In this study, we introduce a speckle-free framework for quantifying local strain fields using a state-of-the-art, zero-shot transformer model, CoTracker3. We utilized a custom-designed, portable isotonic biaxial apparatus compatible with multiphoton microscopy (MPM), to demonstrate this approach, successfully tracking natural bladder lumen textures without artificial markers. Benchmark tests validated the method's high accuracy, achieving a tracking RMSE of less than 1.5 pixels and low strain errors. Our framework effectively captured heterogeneous deformation patterns, despite complex folding and buckling, which conventional, DIC often fails to track. Application to in vitro active bladder contractions in four rat specimens (n=4) revealed statistically significant anisotropy (p<0.01), with higher contraction longitudinally compared to circumferentially. Multiphoton microscopy further illustrated and confirmed heterogeneous morphological changes, such as large fold formation during active contraction. This non-invasive approach eliminates speckle induced artifacts, enabling more physiologically relevant measurements, and has broad applicability for material testing of other biological and engineered systems.
The study's objective is to propose a novel non-invasive method for rapid screening and regular assessment of adolescent idiopathic scoliosis (AIS) through development of a wearable system integrated with multiple inertial measurement units (IMUs) and deep learning models. The system is designed to automatically distinguish between healthy individuals and AIS patients, and subsequently predict the Cobb angle based on continuous temporal kinematic angular sequences acquired during gait. Gait kinematic data were acquired from 124 participants (104 patients with average Cobb angle of 21.62 ± 7.93° and 20 healthy subjects) using a 9-IMU wearable device. Extracted angular features were analyzed to quantify bilateral asymmetry, compare differences across severity subgroups, and evaluate their linear correlation with Cobb angles. A two-stage deep learning framework was implemented with a convolutional neural network (CNN) classification model developed for the rapid screening of scoliosis, and a CNN-Transformer model was designed and compared with other five model architectures to predict Cobb angles from the acquired temporal angular sequences. Scapular kinematics emerged as the most prominent marker of asymmetry, and knee joint kinematics served as the strongest indicator of severity. Meanwhile, angular features of knee, hip and ankle joints demonstrated weak negative linear correlations with Cobb angle. In addition, the scoliosis screening model achieved high predictive performance, with an accuracy of 96.59% and a precision-recall AUC of 0.94 for scoliosis detection. For Cobb angle prediction, the CNN-Transformer model regularized with Gaussian noise during training proved most effective, yielding a mean absolute error of 2.14 ± 0.28° and R 2 value of 0.85 ± 0.03, outperforming other architectural alternatives evaluated in this study. Kinematic analysis of angular data validated the efficacy of the wearable system and effectively captured gait characteristics specific to AIS. The deep learning models accurately distinguished scoliosis patients from healthy cases and predicted Cobb angles using temporal kinematic angular sequences, providing a safe, non-invasive, operator-friendly approach suitable for rapid screening and regular assessment.
Intellectual and developmental disability (IDD) is characterized by impairment of cognitive function that results in daily living limitations. Adults with IDD face barriers to meeting basic needs and must navigate multiple public systems to access services. The housing affordability crisis has contributed to a disproportionate overrepresentation of adults with IDD who experience homelessness. Little evidence has articulated the unique systems- and policy-level issues facing homeless adults with IDD. The purpose of this study was to 1) map the systems with which adults with IDD who experience homelessness must interface in order to access needed services and 2) describe the systems- and policy-level issues impacting service delivery and care coordination of this population. This qualitative study interviewed professionals providing disability, homeless, and/or social care services to adults experiencing homelessness between March and June 2021. Data were analyzed using qualitative content analysis. Participants (n = 18) mostly identified as female (n = 11), the mean age was 44 (13.9), and included clinicians (n = 5), case managers (n = 7), allied professionals (n = 5), and outreach specialists (n = 1). Participants identified five unique systems with which individuals at the intersection of IDD and homelessness frequently interact to access care and services. The theme structurally fragmented care for complex needs captures participants' perceptions of systems-level issues, and the theme policy-driven gaps in care encompasses perceptions of policy-level issues. Homeless adults with IDD exist in a nearly perpetual process of cycling through public systems, and health equity cannot be attained without attention to the structural ableism that is embedded throughout. Findings highlight the importance of moving away from a crisis-driven model of care to a cohesive care ecosystem, and future research is needed to develop and test models of care that eliminate system churn so that all individuals are supported.
Reactive α-dicarbonyls, such as methylglyoxal (MGO), are critical biomarkers of carbonyl stress, yet their real-time monitoring is stifled by a selectivity-biocompatibility paradox. Existing probes either suffer from aldehyde promiscuity, failing to distinguish dicarbonyls from the global lipid peroxidation background, or rely on high-energy, cytotoxic excitation that precludes longitudinal study. Herein, we report the rational design of a dicyano-BODIPY platform engineered to resolve this tension through a precision-tuned acceptor-photoinduced electron transfer (a-PET) mechanism. By employing density functional theory (DFT) as a predictive blueprint, we strategically depressed the BODIPY core HOMO to -6.63 eV, establishing a specific energetic gradient that enforces a robust "off" state until triggered by reversible covalent capture of α-dicarbonyls. This electronically gated design enables longitudinal, visible-light imaging of bidirectional MGO flux in living cells, a feat inaccessible to current irreversible sensors. We further demonstrate the platform's high-fidelity performance in complex biological matrices by mapping dose-responsive MGO burden in murine brain tissue following controlled intracranial perturbation, providing a vital tool for interrogating the role of glyoxal stress in tissue-level pathologies. This work provides a generalizable electronic framework for the development of reversible-covalent sensors capable of monitoring metabolic dynamics in intact biological environments.
Post-traumatic stress disorder (PTSD) remains highly prevalent affecting ~23% of World Trade Center (WTC) responders more than two decades after 9/11. While MRI studies have identified neural differences associated with PTSD, these findings have not translated into improved treatment. We introduce a novel multimodal MRI approach, DAta-driven Network Connectivity Estimate (DANCE), integrating structural and functional magnetic resonance imaging (MRI) to better capture PTSD mechanisms and inform biomarkers. In 96 WTC responders, including 45 with current WTC-related PTSD and 51 without PTSD. We applied graph theory to resting-state functional MRI to identify functional hubs via eigenvector centrality and identified divergence between groups using partial least squares discriminant analysis (PLS-DA). From diffusion MRI, we reconstructed five anatomical tracts (i.e., streamlines) in the temporal lobes. Using DANCE, we quantified the differential distribution of streamlines of the reconstructed tracts connecting the functional hubs. We then tested whether WTC exposure duration moderated associations between PTSD and DANCE indices. Responders with PTSD showed altered centrality in nine functional hubs (AUC=0.75 (0.651-0.847)) including bilateral anterior inferior temporal gyrus, right superior parietal lobule, right anterior parahippocampal gyrus, right anterior/posterior superior temporal gyrus (STG), right caudate nucleus, left amygdala and brainstem. Connectivity differences emerged in four tracts: hippocampus, parahippocampus, inferior and superior temporal gyri (STG). DANCE differed in the inferior fronto-occipital fasciculus (IFOF), medial (IFLmed) and lateral (IFLlat) components of the inferior longitudinal fasciculus and in the middle longitudinal fascicle (MdLF). WTC exposure duration significantly moderated the association between PTSD and DANCE values in the IFLmed, right posterior STG (p= 0.035). Our novel DANCE approach revealed converging functional and anatomical connectivity alterations uniquely associated with PTSD in WTC responders and offers compelling evidence for distinct neurobiological signatures of the disorder. These findings significantly advance our understanding of PTSD pathophysiology and highlight potential biomarkers for diagnosis and targeted intervention.
Dietary fibers can stimulate endogenous glucagon-like peptide-1 (GLP-1) secretion through microbial fermentation and gut hormone signaling, potentially enhancing satiety and supporting weight management. Given the growing interest in non-pharmacological strategies to complement or support tapering of GLP-1 receptor agonist therapy, a structured overview of the human evidence is needed. A pre-registered scoping review was conducted using PubMed, Scopus and Cochrane Central. Randomized controlled trials in adults assessing circulating GLP-1 concentrations and satiety following supplementation with a single, well-defined dietary fiber were included. Fiber types were categorized based on structural characteristics. Outcomes were summarized qualitatively across fiber categories. In total, 1049 papers were screened and 49 publications comprising 52 studies (total n=1,085 participants; median sample size per study=19) were included. Most studies were acute interventions (71%) and conducted in Western populations. Studies reporting increased GLP-1 showed a non-significant tendency to also report increased satiety (OR = 2.95, 95% CI: 0.87-9.98). Dextrins stood out as one of the few fiber categories showing robust effects on both GLP-1 (4 positive studies) and satiety (5 positive studies). Other fibers, such as β-glucans and mannans, showed more uniform effects on satiety or GLP-1, respectively, but did not consistently affect both outcomes simultaneously. Although these findings identify dextrins as a promising dietary fiber candidate for future research, the evidence remains constrained by small sample sizes, short interventions, and substantial heterogeneity. Longer-term studies in free-living conditions, including periods of GLP-1 receptor agonist tapering, are needed to capture microbiota adaptation and generate robust real-world evidence. https://osf.io/cnw4e/overview.
Reliable individual identification is essential for long-term tracking and reproducible laboratory animal studies. In amphibians, invasive marking methods like tags and dye injections can cause welfare concerns and may be unreliable because of tag loss or migration. We developed and evaluated a non-invasive identification system for 25 adult Pleurodeles waltl using smartphone-captured images and the pre-trained convolutional neural network EfficientNetV2. To determine the most informative imaging region, separate models were trained and tested using head and whole-body images. The head-image model achieved 95.3% accuracy on the independent test dataset (macro F1-score = 0.946; Cohen's kappa = 0.951), markedly outperforming the whole-body model (56.6% accuracy). Grad-CAM visualization showed that the model primarily focused on dorsal head spot patterns, indicating that these markings are more informative for individual recognition than whole-body patterns. Because this method requires only a smartphone and a trained model, it can be implemented without specialized marking equipment. This approach enables accurate individual identification, while avoiding invasive marking and therefore supports both animal welfare and research reproducibility. It may provide a practical basis for standardizing individual identification in future amphibian research.
Health behaviours (e.g. exercise/diet/alcohol/smoking) are a major public health concern and traditional complementary and integrative medicine providers such as traditional acupuncturists could make an important contribution. We explored the support for health behaviour change provided to and experienced by traditional acupuncture patients to understand the ways in which traditional acupuncturists may contribute to their patients making changes to their health behaviours. Longitudinal qualitative research methods were used to capture experiences as they change over time with nine patient‒practitioner dyads (8 female patients and 1 male, aged 23-55, and 3 female traditional acupuncturists, aged 35-63). Consultations were audio-recorded and in-depth qualitative interviews were conducted with patients (twice) and traditional acupuncturists (once). Experiences were explored with primary focus on the patients; data were coded both deductively and inductively and themes were produced using reflexive thematic analysis. Patients at different stages of readiness to change need different types of support-those overwhelmed by symptoms/stress needed help to gain control, patients more ready to change needed help sustaining motivation. Acceptance of lifestyle/behaviour advice was related to establishing a cycle of trust, in practitioner, explanations and treatment. Decisions to enact behaviour change were based partly on changes in feelings/mood as well as reasoned understanding (e.g. of behavioural contribution to symptoms). Patients' response to behaviour change support in traditional acupuncture may vary according to their symptom control and stress. For some, support for symptoms/stress may need addressing before any behaviour change intervention. Therapeutic trust, coherent explanations and positive feeling from treatments were important in supporting change. These findings link to concepts from the trans-theoretical model and stages of change, social cognitive theory, and the common sense model of self-regulation and dual process theories of behaviour.
Clinical prognostic models for nasopharyngeal carcinoma (NPC) treated with intensity-modulated radiotherapy (IMRT) with or without chemotherapy remain insufficient to capture tumour heterogeneity. We investigated whether computed tomography (CT)-based signatures add prognostic value for overall survival, progression-free survival, local control and distant control in NPC patients. The study population consisted of 1360 patients with stage I-IVa NPC treated with (chemo)IMRT (2013-2017). Radiomic and deep-learning features were analysed with twelve clinical variables. Radiomic models were built using bootstrap resampling feature selection and multivariable Cox regression; deep-learning models used 3D ResNet-18 or DenseNet-121. Models were evaluated on an internal hold-out test set (n = 409; training set n = 951) with the concordance index and compared against clinical-only reference models. Decision curve analysis was used to assess clinical utility. Adding radiomic primary tumour features (Neighbouring Gray Tone Difference Matrix - coarseness) improved local control concordance index from 0.51 to 0.60 (p = 0.02). A DenseNet-121 combining clinical data with composite primary tumour and lymph node masks achieved the highest distant control (0.68 vs 0.66, p = 0.01). For overall survival and progression-free survival, the improvements were not significant. Decision curve analysis demonstrated net benefit of the DenseNet-121 distant control model over treat-all and treat-none strategies at threshold probabilities of 10-25%. Incorporating CT-based radiomic and deep-learning features into prognostic models significantly improved prediction of local and distant control in NPC, supporting their potential as imaging biomarkers for refined risk stratification.
Magnetic Resonance Imaging (MRI) derived brain age varies substantially between individuals, but it remains unclear whether early deviations from normal brain ageing precede future cognitive decline and whether they provide predictive value beyond conventional MRI measures. Here, we investigated whether MRI-derived brain age gap estimation (BrainAGE) identifies early structural brain ageing differences among cognitively normal individuals who later develop mild cognitive impairment (MCI) or dementia. We analysed longitudinal structural MRI data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and replicated the main findings in the population-based Kuopio Osteoporosis Risk Factor and Prevention Study (OSTPRE). Individuals who later converted to MCI or dementia had higher BrainAGE values several years before diagnosis and, in ADNI, showed steeper longitudinal increases than stable individuals. Elevated BrainAGE values were also associated with increased risk of future conversion to MCI in cognitively healthy individuals and faster subsequent memory decline. Cross-sectional differences and the association between BrainAGE and risk of future conversion were replicated in OSTPRE. Importantly, adding BrainAGE to models including demographic, APOE4, cognitive, and MRI-derived measures consistently improved prediction of future cognitive outcomes, with the greatest benefit observed for individuals who converted after longer follow-up. These findings show that structural brain ageing begins to diverge years before the onset of MCI. BrainAGE captures this early divergence, providing complementary information beyond conventional structural MRI measures that may improve the early identification of cognitively normal individuals at increased risk of future cognitive decline when integrated with other biomarkers.
Wearable sensor-based gait analysis has attracted increasing attention for diverse applications in industrial fields, such as healthcare, rehabilitation, sports science, and continuous monitoring for worker injury prevention. Among gait-related signals, ground reaction force (GRF) is a key indicator for understanding human locomotion and is conventionally measured using force plates or instrumented treadmills. However, such systems are costly and limited to controlled laboratory environments, which restrict their use in daily-life settings. Wearable insole sensors provide a more portable alternative for GRF acquisition, but their measurements are often noisy and less reliable, making accurate estimation challenging. Although deep learning models have shown promise in improving GRF estimation from wearable sensor data, high-performing models are often computationally expensive and difficult to deploy on resource-constrained devices. Moreover, since gait signals exhibit phase-specific temporal structures, effective estimation requires the model to develop a semantically meaningful understanding of the gait sequence rather than merely matching signal values. In this regard, human-interpretable semantic information can provide useful guidance for modeling such structured temporal dynamics. To address these issues, we propose a Gait-Semantic Relational Knowledge Distillation (SRKD) framework for GRF estimation using wearable insole sensor data. The proposed method introduces gait-phase-aware semantic guidance into the teacher-student distillation process by decomposing intermediate representations into phase-specific components and aligning them with corresponding textual semantic representations. This phase-aware distillation framework uses textual semantics as anchors to bridge the teacher-student representational gap, enabling the student to effectively learn the structured temporal relational knowledge underlying gait dynamics. We evaluate SRKD under several experimental configurations, including different teacher-student architectures, temporal window lengths, text encoders, and human-refined semantic descriptions. The experimental results show that SRKD generally improved the estimation performance of the evaluated lightweight student models compared with the considered baseline methods. Qualitative analyses also indicate that the predicted GRF signals more closely captured both the overall waveform and phase-dependent temporal variations in the evaluated settings. Furthermore, the results obtained using human-refined descriptions suggest that human-guided semantics can enhance the effectiveness of knowledge distillation by providing more task-relevant semantic guidance. Further validation across broader populations, sensor configurations, and real-world environments is required to assess the generalizability and practical applicability of the proposed framework.
Traumatic brain injury (TBI) severity is typically classified using clinical indices that may have limited prognostic value. Objective measures of olfactory function depend on sensory, limbic, and memory networks and may provide a more specific marker of injury-related neural dysfunction. Seventy-nine individuals with TBI (48 mild, 31 moderate-to-severe) and 59 healthy controls completed an olfactory battery, including tests of odor percept identification (OPID9, OPID18), odor discrimination (OD10), and odor memory (POEM) ∼4.04 (4.45) years after their most recent TBI. General linear models examined associations between olfactory outcomes and TBI severity, adjusting for age, age², sex, and education. Additional models examined the relationship between loss of consciousness (LOC) and olfactory functioning. The severity of the most recent TBI was significantly associated with all olfactory outcomes, after adjusting for age, sex, and education. Compared with controls, participants with moderate- to-severe TBI showed lower OPID9, OPID18, POEM, and OD10 performance, while participants with mild TBI showed lower OPID18 and POEM performance. LOC was associated specifically with odor memory in these models, as participants with prolonged (> 30min) LOC or LOC of unknown duration had lower POEM scores than those with no LOC. In TBI-only models, LOC remained associated with POEM after adjustment for TBI severity, whereas TBI severity was not associated with POEM after LOC was included. Long term olfactory functioning is sensitive to TBI severity, with generalized impairments across odor identification, discrimination, and odor memory. LOC characteristics appear especially relevant to odor memory, suggesting that olfactory memory may capture injury-related features beyond TBI severity classification alone.
Reaching the Landsberg limit for radiative energy conversion requires controlling not only spectra but also the direction in which radiative entropy is transported. We develop a finite-stage nonreciprocal thermal-circulator framework for positive illumination (solar conversion) and negative illumination (radiative-cooling work extraction). The ideal theory is expressed in terms of open-system exergy and a minimal port-network picture: a three-port circulator with a cold termination acts as a two-port thermal isolator, and cascaded isolators enforce directed adjacency between effective photon reservoirs. In response to the central practical limitation of nonreciprocal radiative cooling, emphasized recently by Liu et al. we recast the problem as a benchmark map rather than a claim of immediate device-level cooling enhancement. The reported work output values (in unit of W m-2) are hemispherical-equivalent upper envelopes; experimentally relevant output is reduced by captured free-space etendue, coupling efficiency, finite isolation ratio, finite termination temperature, atmospheric transmission, and out-of-window loss. We therefore introduce explicit sensitivity factors for isolation, termination, and captured etendue and ask how good a nonreciprocal system must be before the finite-stage advantage remains measurable. Finite-stage calculations give temperature and wavelength targets for 3-5 stages, while the practical benchmark identifies the isolation, termination, and coupling requirements needed to make those targets consequential. This formulation links ideal Landsberg-type limits to testable design requirements for nonreciprocal thermal photonics.