SEARCH · Search PubMed

Results for “regulatory network”

Search indexed PubMed citations on genomics, clinical trials, systematic reviews and public health. Explore titles, authors and supplied subject terms, then open the PubMed record.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

312 records · Page 6Linked to original sources

PubMed2026

Efficacy, acceptability, and related outcomes of pharmacological interventions for acute bipolar mania: a systematic review and dose-related network meta-analysis across different age groups.

BACKGROUND: Acute bipolar mania carries negative social and economic consequences. We investigated the comparative efficacy/response/acceptability of pharmacological interventions for acute bipolar mania, considering dose effects across different age groups. METHODS: We conducted a network meta-analysis (NMA) to search for randomized controlled trials (RCTs) comparing pharmacological interventions with one another or placebo in acute bipolar mania patients, indexed in PubMed/MEDLINE, Embase, Web of Science, and Scopus (from inception through 2025.12.24). Co-primary outcomes were change in manic symptoms/response/and acceptability. Tolerability/remission and rate of adverse events were secondary outcomes. Confidence-In-Network-Meta-Analysis was likewise appraised. RESULTS: 113 RCTs, encompassing 49 distinct treatment combinations, included 20,666 participants. Sensitivity analysis retaining only low-risk-of-bias studies and excluding outliers for possible effect modifiers indicated that risperidone 3 mg/day(SMD = -7.57;95%C.I. = -8.25;-5.85); tamoxifen 160 mg/day(SMD = -1.73;95%C.I. = -2.32;-1.13); rivastigmine 3 mg/day(SMD = -1.13;95%C.I. = -1.06;-0.58); haloperidol 30 mg/day(SMD = -0.96;95%C.I. = -1.25;-0.75); valproate 750 mg/day(SMD = -0.76;95%C.I. = -1.48;-0.58); tamoxifen 40 mg/day(SMD = -0.75;95%C.I. = -1.41;-0.59); celecoxib 400 mg/day(SMD = -0.74;95%C.I. = -1.20;-0.38); paliperidone extended-release 12 mg/day(SMD = -0.62; 95%C.I. = -0.91;-0.32); olanzapine 15 mg/day(SMD = -0.59;95%C.I. = -0.60;-0.38); olanzapine 20 mg/day(SMD = -0.52;95%C.I. = -0.66;-0.38); risperidone 4 mg/day(SMD = -0.53;95%C.I. = -0.76;-0.29); allopurinol 600 mg/day(SMD = -0.54;95%C.I. = -0.67;-0.22); cariprazine 12 mg/day(SMD = -0.49;95%C.I. = -0.66;-0.33); risperidone 4.2 mg/day(SMD = -0.46;95%C.I. = -0.75;-0.17); lithium 1500 mg/day(SMD = -0.42;95%C.I. = -0.57;-0.28); ziprasidone 160 mg/day(SMD = -0.49;95%C.I. = -0.68;-0.31); asenapine 20 mg/day(SMD = -0.38;95%C.I. = -0.53;-0.22); haloperidol 8 mg/day(SMD = -0.34;95%C.I. = -0.63;-0.05); aripiprazole 15 mg/day(SMD = -0.33;95%C.I. = -0.61;-0.06) outperformed placebo. Ziprasidone 160 mg/day, celecoxib 200 mg/day, asenapine 20 mg/day, and asenapine 10 mg/day proved more efficacious than placebo in children. No statistically significant differences were reported between treatments and placebo for response/remission/acceptability/tolerability, and manic/hypomanic switch. A meta-regression of efficacy effect sizes against the adapted AMSTAR-Plus content scores showed that larger SMDs were associated with lower AMSTAR scores, indicating lower study quality, warranting further caution for such large efficacy estimates. CONCLUSIONS: Our findings are consistent with previous NMAs and current guidelines, expanding the current knowledge base while concurrently appraising different drugs, doses, and age groups.

Humans↗

PubMed2026

Urban stormwater infrastructure as a microplastic superhighway: a critical review of transport dynamics, modelling, and mitigation across pavements and drainage networks.

This review examines the transport, fate, modelling, and mitigation of Microplastics (MPs) in urban stormwater infrastructure, with emphasis on pavements, runoff pathways, micro-drainage, and macro-drainage systems. Following a systematic review approach, more than 1000 records were screened and approximately 50 core studies were retained when they addressed urban stormwater or drainage-related MP transport with adequate methodological reporting; marine-only studies and biological-effect studies without direct relevance to transport processes were excluded. The evidence shows that stormwater systems function not merely as passive conduits but as dynamic reactive transport systems with temporary storage, where particle mobilisation, sedimentation, resuspension, and temporary retention regulate MP export. Road surfaces, especially high-traffic areas, are major reservoirs of tyre wear, road-marking, atmospheric, and litter-derived particles that are rapidly mobilised during rainfall. Conventional grab sampling may underestimate MP loads, which in some cases exceed treated wastewater effluent loads by up to six-fold. Drainage structures such as manholes can immobilise up to 17.3% of near-neutrally buoyant particles, while biofouling and aggregation may shift buoyant polymers from wash-load to bedload. Mitigation systems, including permeable pavements, bioretention, wetlands, and technical inserts, can achieve high removal of coarse MPs, but performance declines for fine particles below 100 µm. The review highlights the need for standardised flow-proportional sampling, physically informed modelling, and treatment-train strategies targeting both surface sources and in-network storage.

Microplastics↗

PubMed2026

Infra-low-frequency neurofeedback alters EEG network efficiency: exploratory evidence from healthy volunteers.

Infra-Low-Frequency Neurofeedback (ILF-NFB) combines classic frequency-band (FB) and infra-low-frequency (ILF) EEG components in implicit training protocols and is increasingly applied in clinical contexts. Yet, the neurophysiological mechanisms underlying ILF-NFB remain to be further elucidated. In this randomized, sham-controlled and double-blind study, we explored the online impact of a one-session ILF-NFB application on EEG correlates in healthy participants (39 analyzed datasets). Continuous 31-channel EEG was recorded during verum and sham feedback in a double-blind, randomized crossover design. In this exploratory analysis approach, functional connectivity was estimated using the debiased weighted phase-lag index (dwPLI) and analyzed with graph-theoretical measures. The results revealed higher global efficiency during verum compared to sham in the Beta1 band (12-15 Hz), reaching significance in the primary comparison but not surviving Bonferroni correction across the five tested bands; block-wise follow-ups showed a significant verum-sham difference in the first half of the neurofeedback session and a directionally consistent pattern in the second half. The Condition × Block interaction was not significant. No consistent differences were observed in other frequency bands, nor for betweenness centrality. While preliminary, these exploratory results point to possible network-level effects during ILF-NFB and motivate further confirmatory work in extended training protocols and clinical populations.

Humans↗

PubMed2026

Cerebellar iTBS enhances gait adaptation by modulating cortical sensorimotor network dynamics: a randomized controlled trial.

Gait adaptation enables individuals to maintain locomotor stability under persistent perturbations. Although the cerebellum is critical for sensory prediction error-based (SPE) adaptation, how cerebellar neuromodulation reshapes cortical sensorimotor networks to enhance gait adaptation remains unclear. This study investigated the behavioral effects and underlying cortical neurodynamic mechanisms of cerebellar intermittent theta-burst stimulation (iTBS) on gait adaptation. Thirty-two healthy adults received either active or sham cerebellar iTBS. Participants performed a split-belt treadmill adaptation task before and after intervention. Cortical responsiveness was evaluated using TMS-evoked EEG over primary motor cortex (M1), while resting-state EEG was analyzed to assess spectral power and directional functional connectivity. Compared to sham, cerebellar iTBS significantly enhanced gait adaptation, evidenced by a faster adaptation rate (p = 0.035) and enhanced Early Adaptation SLS (p = 0.011), without altering initial perturbation responses or post-adaptation outcomes. The iTBS increased TMS-evoked α (p = 0.031) and γ (p = 0.022) power in M1, while the α power was correlated with faster adaptation (r = 0.526, p = 0.002). Furthermore, iTBS strengthened PPC-to-M1 directed connectivity in the β (p = 0.025) and γ (p = 0.013) bands. Enhanced parieto-motor directionality were positively associated with adaptation rate (β: r = 0.515, p = 0.003; γ: r = 0.463, p = 0.009). These findings suggest that cerebellar iTBS facilitates gait adaptation by modulating cortical responsiveness and directional sensorimotor network connectivity, providing multi-level neurodynamic evidence for the cerebello-cortical modulation during gait adaptation and offering a strong physiological rationale for targeted neuromodulation in gait rehabilitation strategies.

Humans↗

PubMed2026

Neurocorrelates of nocturnal enuresis in pre-adolescent children.

INTRODUCTION: Nocturnal enuresis (NE) is a common neurodevelopmental condition, yet its underlying neural mechanisms remain unclear. This study leverages the large-scale Adolescent Brain Cognitive Development (ABCD) dataset to identify structural and functional brain correlates associated with active symptoms and the resolution of bedwetting. METHODS: Using cross-sectional data from 3472 participants aged 9-10 years, children were categorized into three groups: active nocturnal enuresis (ANE, n = 225), history of nocturnal enuresis (HNE, n = 1171), and healthy control groups (CG, n = 2076). Multimodal neuroimaging protocol evaluated macrostructural properties via structural MRI (sMRI), microstructural white matter integrity via diffusion MRI (dMRI), and functional connectivity via resting-state fMRI (fMRI). Group differences were evaluated using linear models within an ANCOVA framework, adjusting for intracranial volume and handedness with False Discovery Rate (FDR) correction. RESULTS: Compared to controls, the ANE group exhibited a significant volume deficit in the right caudate, decreased sulcal depth in the left insula, and lower internal correlation within the Cingulo-Opercular Network (CON). Conversely, the dry HNE group demonstrated significant structural adaptations, including bilaterally larger putamen volumes and increased right caudate volume compared to the ANE group. The HNE group also showed increased microstructural density (decreased mean diffusivity) in the bilateral hippocampus and an increased cortical surface area in the left insula. Both NE groups demonstrated persistently reduced functional coupling within the CON. CONCLUSIONS: Nocturnal enuresis appears to be associated with a potential complex central signaling deficits. Reduced internal correlation within the CON across both active and former bedwetters indicates a potential for impairment in processing internal homeostatic bladder signals during sleep.

Humans↗

PubMed2026

Cytonuclear conflict and reticulate evolution in the Morelloid clade (Solanum, Solanaceae): Insights from genome skimming and network Phylogenomics.

The Morelloid clade (black nightshades) is one of the most strongly supported clades within the megadiverse Solanum genus. It comprises 76 globally distributed, non-spiny herbaceous and suffrutescent species. While often erroneously considered poisonous weeds, several species are economically important as orphan crops. The clade is closely related to tomato and potato but, due to a lack of focused breeding efforts, remains a putative reservoir of genetic diversity for crop improvement. Despite this potential, we lack fundamental knowledge on the evolution of the Morelloid clade. The group includes polyploid species with unknown parental origins-likely reflecting reticulate processes such as hybridization, introgression, and associated backcrossing events. Prior analyses have been unable to disentangle these processes, leaving the mechanisms underlying reticulate evolution in the Morelloid clade poorly understood. Here, we use genome skimming to produce a well-supported maximum likelihood plastid phylogeny from complete circularized plastomes and a coalescent-based species tree from combined Angiosperms353 and conserved ortholog set nuclear markers. Our dataset, composed of previously published data and deep genome skimming from herbarium samples, spans 26 Morelloid species. To investigate phylogenetic discordance, we used a nuclear phylogenetic network, multispecies coalescent simulations, a fused rooted nuclear chloroplast tree, and quantification of nuclear gene tree concordance. We show that incongruence between nuclear and plastid trees is pervasive and cannot be explained by incomplete lineage sorting alone. Instead, our results demonstrate that events consistent with repeated chloroplast capture have shaped the reticulate evolutionary history of the clade, especially among African polyploid and Pan-American diploid lineages.

Phylogeny↗

PubMed2026

The Thyroid-Brain Network: Exploring Inflammation, Immune Mechanisms and Common Triggers in Thyroid-Related Neurological Dysfunction.

Autoimmune thyroid diseases (AITD), including Hashimoto's thyroiditis and Graves' disease, represent the most prevalent endocrine disorders worldwide, affecting hundreds of millions with profound but often under recognized neurological consequences. There are emerging lines of evidence establishing inflammation and immunity as the critical missing link connecting peripheral thyroid dysfunction to central nervous system manifestations. Thyroid hormones function as essential neuromodulators governing neurodevelopment, synaptic plasticity, and cognitive processing through integrated genomic and non-genomic mechanisms, with region-specific cerebral metabolic disturbances correlating with distinct neuropsychiatric symptoms. The immunological perspective reveals that AITD propagates neuroinflammation through convergent pathways: molecular mimicry enabling cross-reactivity between thyroid and neural antigens, cytokine-mediated disruption of neurotransmitter metabolism, HMGB1-driven glial activation, and blood-brain barrier compromise facilitating immune cell infiltration. The thyroid-gut-microbiota axis emerges as a critical mediator wherein dysbiosis perpetuates both thyroid autoimmunity and neuroinflammation through impaired serotonin precursor availability and increased intestinal permeability. Mitochondrial dysfunction represents an energetic common denominator, as thyroid hormone dysregulation directly impairs oxidative phosphorylation, producing region-specific cerebral metabolic disturbances. Simultaneous compromise of monoamine systems, cholinergic signaling abnormalities, and glutamate excitotoxicity creates a particularly toxic neurochemical state in untreated thyroid dysfunction. Common triggers such as psychological stress, gut dysbiosis, and mitochondrial impairment may activate interconnected pathways that simultaneously compromise thyroid and brain function, revealing that these disorders share fundamental mechanistic origins. These insights have been discussed in the current review to enhance the understanding of thyroid-brain function, the core mechanisms and consequences of functional deficits.

Journal Article↗

PubMed2026

Dual Transcranial Direct Current Stimulation Modulates Hierarchical Functional Network Organization in Post-Stroke Cognitive Impairment: A Randomized Controlled Trial.

OBJECTIVE: To evaluate the clinical efficacy of dual transcranial direct current stimulation (tDCS) in patients with post-stroke cognitive impairment (PSCI) and to explore the effects on the hierarchical organization of functional brain networks, ranging from regional synchronization to inter-regional connectivity and global network topology. METHODS: In this randomized, double-blind, sham-controlled trial, 74 PSCI patients received conventional therapy alongside either active dual-tDCS (n = 38) or sham stimulation (n = 36). Active tDCS targeted the dorsolateral prefrontal cortex (DLPFC) via anodal-left/cathodal-right nodes (2.0 mA, 20 min/day, 20 sessions). The primary outcome was the Montreal Cognitive Assessment (MoCA). Secondary outcomes included the Mini-Mental Status Examination (MMSE), Stroop Test (ST), Trail Making Test (TMT), Wechsler Memory Scale (WMS), and Barthel Index (BI). A subgroup of 36 participants (18 per group) underwent resting-state functional magnetic resonance imaging (rs-fMRI) to analyze regional homogeneity (ReHo), functional connectivity (FC), and network topology. Partial correlations assessed the association between neuroimaging alterations and clinical improvements. RESULTS: The tDCS group showed significantly greater improvements in MoCA scores (tDCS: 5.74 ± 2.76 vs. sham: 2.69 ± 2.69; t = 4.799, p < 0.001) as well as in attention and memory domains compared to the sham group. The rs-fMRI changes included increased ReHo in the right middle temporal gyrus (MTG) and the left inferior frontal gyrus (IFG), and reduced FC between the right MTG-left superior frontal gyrus and left IFG-cerebellum (p < 0.05, FWE-corrected). Additionally, small-worldness and global efficiency increased (p < 0.05) with these alterations correlating with clinical recovery. Adverse events were rare and self-limiting. CONCLUSION: Dual-tDCS over bilateral DLPFC safely improves cognitive recovery in PSCI. These clinical gains are associated with rs-fMRI alterations, specifically in regional synchronization, inter-regional connectivity, and global topology, which suggest a potential biomarker for monitoring tDCS efficacy, offering a rationale for precision neuromodulation in stroke rehabilitation.

Humans↗

PubMed2026

Smoking Cue Reactivity in Relation to Uncertain-Threat and Reward-Anticipation Networks: A Coordinate-Based fMRI Meta-Analysis.

BACKGROUND: Smoking is a concerning medical and social problem, yet how the brain links stress to continued smoking is still not well understood. This coordinate-based meta-analysis identified convergent activations for smoking cues, uncertain threat, and reward anticipation, and examined co-activation patterns to clarify whether smoking-cue activity in smokers relates to threat and reward activity in non-addicted controls. METHODS: We conducted a coordinate-based activation likelihood estimation (ALE) meta-analysis of 102 fMRI studies (N = 3,068), including 30 studies on smoking cue reactivity (n = 945), 48 on reward anticipation (n = 1,413), and 24 on uncertain threat processing (n = 1,621). We performed single, conjunction, and contrast analyses, followed by meta-analytic connectivity modeling (MACM) of key regions. RESULTS: Single analysis revealed smoking engaged bilateral ACC (-1.1, 46.2, -1.1), uncertain threat engaged bilateral insula (left = -33.6, 22, 4.3, right = 41.3, 22, 1), reward anticipation activated thalamus (0.9, 1.3, -3.1) and medial frontal gyrus (2.7, 6.6, 52.1). Conjunction and contrast analyses showed shared or unique activation for each task in its respective regions. MACM showed ACC co-activation with thalamus and medial frontal gyrus, while insula co-activated with ACC and inferior frontal gyrus. CONCLUSIONS: Each process converged in a separate region, with no overlap between the smoking-cue map and either the threat or reward map. The ACC nonetheless co-activated with reward-related regions and shared network membership with the threat-related insula. On this basis we hypothesize a shift in motivation from stress-driven reward toward cue-driven craving, to be tested within subjects, and identify candidate neuromodulation targets for preventing stress-precipitated relapse.

fMRI↗

PubMed2026

Pesticide occurrence, transformation, and transport from wastewater treatment plants into stream networks with diverse land uses.

Neonicotinoid insecticides and strobilurin fungicides are detected in many environmental compartments and have been associated with negative environmental and human health implications. Wastewater treatment plants (WWTPs) are often hotspots for introducing such contaminants into the environment. Therefore, the occurrence of strobilurin fungicides, neonicotinoids, and their metabolites at two WWTPs with varying land uses and population sizes was investigated. Polar organic chemical integrative samplers were deployed in WWTP influent and effluent and placed upstream and downstream of the effluent mixing zone for 2 weeks in April and July 2022. Biosolids were also collected at each time point. Neonicotinoids were detected with the highest frequency (68%), followed by strobilurin fungicides (49%) and neonicotinoid metabolites (31%). Time-weighted average concentrations for influent/effluent ranged from 85.2 ± 87.8 to 409.4 ± 74.5 ng/L. Pesticide concentrations, specifically the metabolites, typically increased from influent to effluent, resulting in effluent having higher pesticide loads than influent. Pesticide concentrations varied between the upstream and downstream monitoring locations by analyte, with WWTP samples in the highly developed region having significantly higher concentrations of pesticides and less variation by monitoring period. Chronic ecotoxicity benchmarks for freshwater invertebrates for imidacloprid were surpassed in treated effluent at both WWTPs in July and in the downstream monitoring location in the heavily developed area. Findings support the need for further exploration of pesticide contributions from WWTPs to river systems, specifically related to metabolite contributions to downstream streams and their effects on aquatic environments.

Water Pollutants, Chemical↗

PubMed2026

Genome-Wide Impact of Human DBR1 Depletion on RNA Processing Networks Reveal a Connection Between Pre-mRNA Splicing, mRNA Surveillance and Stress Granule Dynamics.

The RNA lariat debranching enzyme DBR1 is essential for intron turnover and RNA metabolism, yet its broader impact on transcriptome regulation remains incompletely defined. To elucidate the consequences of DBR1 depletion, we performed transcriptome-wide RNA sequencing of DBR1-knockdown and wild-type HEK293 cells. Differential expression analysis revealed widespread perturbations in pathways linked to RNA splicing, mRNA surveillance, translational control, and stress-granule biology. Many of the most significantly altered transcripts encode splicing factors and RNA quality-control components, underscoring DBR1's influence on post-transcriptional regulation. Alternative splicing analysis showed changes across multiple event types, with exon skipping accounting for >50% of events, followed by mutually exclusive exons, alternative 5' and 3' splice sites, and retained introns, indicating that DBR1 depletion induces pervasive splicing defects. Direct spliceosome inhibition using isoginkgetin (blocks tri-snRNP recruitment) and pladienolide B (targets SF3B1) reproduced the DBR1-KD mis-splicing patterns of cell signaling genes and factors involved in RNA metabolism, supporting a functional link between DBR1 activity and alternative splicing. Notably, DBR1 knockdown revealed a subset of transcripts that are both NMD-sensitive and enriched within stress granules. Consistent with this observation, G3BP1 immunopurification and confocal microscopy further support a role for DBR1 and UPF1 in stress-granule dynamics, suggesting that these factors may participate at distinct stages to influence mRNA fate under stress conditions. Together, these findings indicate that DBR1 functions beyond lariat RNA turnover as a common regulator of RNA processing, transcriptome stability, and stress granule homeostasis, revealing intricate crosstalk between RNA splicing and RNA quality control pathways in human cells.

Humans↗

PubMed2026

ORBIT: Oncogenic Representation Learning via Bi-Prototype Contrastive Learning in Hyperbolic Space for cancer driver gene identification.

Accurate identification of cancer driver genes is crucial for precision oncology but remains challenging due to the complexity of integrating heterogeneous data and modeling dynamic biological systems. To address these limitations, we propose ORBIT (Oncogenic Representation Learning via Bi-Prototype Contrastive Learning in Hyperbolic Space). Our framework synergistically fuses multi-omics profiles with functional network data using a context-adaptive graph reweighting mechanism to capture cancer-specific dynamics. The model employs a bi-prototype contrastive learning strategy within hyperbolic space, which aligns gene representations around distinct driver and non-driver semantic anchors while preserving the intrinsic hierarchy of biological networks. Comprehensive evaluations demonstrate that ORBIT achieves highly competitive stability in pan-cancer analysis while consistently outperforming state-of-the-art methods in cancer-specific predictions. Furthermore, functional enrichment analysis confirms that the model effectively segregates core cancer pathways, and drug sensitivity profiling validates the clinical relevance of the identified drivers. By integrating hyperbolic geometry with context-adaptive learning, ORBIT offers a robust and interpretable paradigm for precision medicine. The source codes and datasets are publicly accessible at https://github.com/spcho-dev/ORBIT.

Humans↗

PubMed2026

Delphi study robot consenso: Strategies for the implementation of robotic surgery in general surgery in the Spanish hospital network.

INTRODUCTION: The implementation of robotic surgery in public hospitals presents multiple logistical, educational, and organizational challenges. In the absence of unified guidelines, a national consensus is required to optimize its safe and efficient adoption. This study aimed to establish a set of consensus-based and measurable recommendations for the implementation of robotic surgery programs in hospitals within the Spanish National Health System, based on the experience of centres with established robotic programs and intended to serve as guidance for hospitals that are initiating or planning their implementation. METHODS: A national Delphi study was conducted with the participation of robotic surgery experts from 26 public hospitals. The expert panel was composed exclusively of digestive surgeons with experience in robotic surgery. Three iterative rounds of expert panel evaluation were conducted between March 2024 and March 2025. The questions were grouped into five thematic blocks. Consensus was defined as an agreement level of ≥66.7%. Kendall's W coefficient was used to assess concordance. RESULTS: High levels of consensus were achieved on key aspects related to infrastructure, structured training, cost evaluation, and quality assurance mechanisms. Areas of disagreement were also identified, such as the need for a dedicated anaesthesiologist, purchase of accessory instruments during the initial phase, and official accreditation pathways. CONCLUSIONS: This study provides a guideline for developing a national robotic surgery strategy focused on patient safety, program sustainability, and standardized training of surgical teams. These recommendations can guide hospitals at different stages of robotic technology adoption. Given that the consensus was reached from an exclusively surgical perspective, the recommendations focus on patient safety, program sustainability, and standardized training of the surgical team, and should be interpreted in an adaptable manner according to each centre's context, case volume, and available resources.

Cirugía Asistida por Robot↗

PubMed2026

Systematic evaluation of one-dimensional-to-two-dimensional near-infrared spectroscopy transformations with deep learning for quantifying coconut sap adulteration.

Near-infrared (NIR) spectroscopy have limitations when combined with deep learning (DL) algorithms because they rely on low-dimensional datasets. Therefore, we investigated the potential of transforming one-dimensional (1D) NIR spectra into two-dimensional (2D) spectrograms using synchronous and asynchronous techniques and the continuous wavelet transform (CWT) and their effectiveness by integrating with DL for detecting adulteration in coconut sap. NIR spectra (12,500-4000 cm-1) were collected from binary mixtures (0%-100%;w/w). The performance of all DL (convolutional neural networks-CNN, AlexNet and ResNet) models was compared with that of partial least squares (PLS). The models were ranked in the mentioned order based on their performances: 2D-CWT > 2D-asynchronous > 2D-synchronous > 1D/2D-PLS. The important features of the best model can be explained and visualized using gradient-weighted-class-activation-mapping. The findings highlight that the 1D-to-2D NIR data transformation combined with DL is a highly robust approach because it addresses the feature representation gap in NIR data and effectively captures the spatial-spectral correlations.

Spectroscopy, Near-Infrared↗

PubMed2026

Quantitative assessment of the fingerprint evidential value using machine learning.

Fingerprints as physical evidence have long supported criminal investigation and adjudication. In practice, however, fingerprint identification relies mainly on examiners' experience. Furthermore, expert opinions tend to be categorical, even though the opinions with the same conclusion could differ substantially in evidential strength. To quantitatively assess fingerprint evidential value, this study proposes a machine learning-based framework as an interpretable decision-support tool. A lightweight residual one-dimensional convolutional neural network was constructed, incorporating channel recalibration and a similarity-driven attention mechanism to learn adaptive contribution weights for different matched minutiae (minutiae for short). Controlled experiments revealed that the predicted evidential value increased with the number of minutiae and was significantly influenced by the quality of minutiae. With 10 minutiae, the mean predicted scores were 4.49, 7.00, and 9.09 for blurred, moderately blurred, and clear minutiae, respectively. Multiple regression analysis indicated that replacing a pair of blurred minutiae with a pair of clear minutiae increased the score by 0.492, whereas replacing it with a pair of moderately blurred minutiae increased the score by only 0.216. By mapping predicted scores to graded levels of evidential strength, the framework contributes to a paradigm shift from categorical expert opinions to graded ones, helping courts evaluate fingerprint evidence more scientifically.

Humans↗

PubMed2026

Implicit and explicit statistical learning in reading: Evidence from a randomized controlled-learning study and computational modeling.

A key challenge in reading acquisition is understanding how learners extract the complex probabilistic mappings between print, meaning, and sound. Statistical learning (SL) theory offers a mechanistic account of how such mappings are acquired, whether implicitly through exposure or explicitly through instruction. We conducted a randomized controlled-learning study in Chinese, a writing system characterized by multiple sub-lexical regularities linking orthography, semantics, and phonology. Ninety-five 2nd-3rd graders with or at risk for dyslexia were randomly assigned to one of three groups: an implicit-SL training group exposed to repeated lexical and sublexical orthography-semantics-phonology associations, an explicit-SL training group receiving the same input plus explicit instruction on the sublexical print-sound mapping, and a no-SL control group. Both SL groups outperformed controls on the characters they were trained on, as well as on untrained characters that required generalization. However, only the explicit group demonstrated abstraction of print-sound mapping to novel items. Neural network simulations further revealed distinct mechanisms supporting implicit and explicit SL, consistent with a dual-system account of reading acquisition. Together, these findings (1) clarify how implicit and explicit learning distinctly support the discovery of statistical structure in written language and (2) underscore the implicit-explicit dual learning mechanism underlying reading acquisition.

Humans↗

PubMed2026

Playing with Fire, Losing the Drive: Bidirectional Links Between Problematic Smartphone Use and Grit Dimensions.

This study examined bidirectional longitudinal associations between grit dimensions (consistency of interest [CI] and perseverance of effort [PE]) and problematic smartphone use (PSU) and tested cognitive flexibility as a mediating mechanism. A sample of 1,641 Chinese university students (55.2 percent female; Mage = 20.1 years) completed measures at two time points 6 months apart. A four-variable cross-lagged panel model revealed that CI and PSU negatively predicted each other over time, whereas PSU unidirectionally predicted decreased PE. Cognitive flexibility partially mediated the PSU-to-PE pathway (indirect effect = -0.004, 95 percent bootstrap CI [-0.010, -0.0001]). Competing models analysis confirmed this directionality: the forward mediation (PSU → cognitive flexibility → PE) was significant, whereas the reverse was not. These findings demonstrate that grit dimensions exhibit distinct longitudinal patterns with PSU and identify cognitive flexibility as a cognitive mechanism through which PSU specifically undermines effort persistence. Implications for dimensional approaches to grit and targeted interventions are discussed.

Humans↗

PubMed2026

Reliability-aware hierarchical learning for Chagas disease screening from 12-lead ECGs: tackling label uncertainty and class imbalance.

Objective.Chagas disease, a neglected tropical disease (NTD) with significant cardiovascular impact, remains underdiagnosed in resource-limited regions. Electrocardiogram (ECG) screening offers a low-cost tool for detecting cardiac involvement, yet algorithm development is challenged by label noise, data scarcity, and the latent nature of infection. This study proposes a robust ECG-based screening framework that explicitly addresses these constraints.Approach.We introduce aReliability-Aware Hierarchical Learningstrategy that calibrates supervision according to data provenance, prioritizing serology-confirmed labels over noisy self-reports. To mitigate data scarcity, we compare a specialized convolutional neural network (CNN) trained from scratch with a transfer learning approach based on a Spatio-Temporal ECG foundation Model (FM). Performance is evaluated across varying data scales, and the representation structure is analyzed to interpret model behavior.Main results.On the official hidden test set of the George B. Moody PhysioNet/Computing in Cardiology Challenge 2025, our approach achieved a Challenge Score of 0.163. We observe that while the specialized CNN performs competitively in data-rich regimes, the FM exhibits superior robustness in extreme low-resource settings. Furthermore, performance reaches a plateau imposed by underlying disease physiology. Bimodal score distributions suggest that models distinguish established cardiomyopathy from indeterminate infection, which remains electrophysiologically indistinguishable from healthy controls.Significance.These findings clarify both the potential and intrinsic limits of ECG-based AI screening for NTD-associated cardiac involvement. Reliability-aware supervision and data-efficient transfer learning provide a practical framework toward scalable and clinically meaningful ECG screening systems in resource-constrained environments.

Humans↗