Tag: pubmed

Thromb Res. 2026 Jun 24;264:109759. doi: 10.1016/j.thromres.2026.109759. Online ahead of print.

ABSTRACT

BACKGROUND: Vitamin K antagonists (VKA) are the recommended treatment for thrombotic antiphospholipid syndrome (TAPS), but it remains uncertain whether direct oral anticoagulants (DOAC) could serve as an alternative to VKA in a subgroup of low-risk TAPS patients with venous thromboembolism (VTE).

OBJECTIVES: To compare the incidence of thrombosis recurrence between DOACs and VKAs in TAPS patients with VTE as index event.

METHODS: In this retrospective cohort study, we included adults with TAPS and index VTE treated with a DOAC or VKA between January 2013 and March 2023. Outcomes were recurrent thrombosis, defined as arterial thromboembolism (ATE) or VTE, and bleeding. Cox regression with time-varying covariates was applied, adjusted for age and sex and stratified by laboratory risk profile according to the 2023 ACR/EULAR criteria.

RESULTS: We included 277 patients (mean follow-up 3.9 ± 2.8 years); 87% single positive. Recurrent thrombosis occurred in 26 patients (9.4%). The risk of the combined ATE/VTE outcome was higher in the DOAC group than in the VKA group, but not statistically significant (aHR 1.90, 95% CI 0.81-4.49). When analyzed separately, DOAC use was associated with more arterial events (1.34 vs. 0.51 per 100 person-years; HR = 3.72 95% CI 1.04-13.29). Major bleeding occurred in 4.0% of patients, with similar rates between treatments (aHR 0.67, 95% CI 0.34-1.31).

CONCLUSION: In TAPS patients with index VTE, DOACs were associated with more arterial recurrences, with no significant difference in the combined ATE/VTE outcome. Interpretation is limited by few events and retrospective design.

PMID:42456240 | DOI:10.1016/j.thromres.2026.109759

Neurotherapeutics. 2026 Jul 15;23(4):e00967. doi: 10.1016/j.neurot.2026.e00967. Online ahead of print.

ABSTRACT

Thalamic neuromodulation has emerged as a promising therapy to reduce seizures in patients with drug-resistant epilepsy who are not candidates for resective or ablative surgical treatments. This advance creates therapeutic opportunities for patients with seizure foci that are generalized, poorly localized, multifocal, or in eloquent cortex. However, optimal treatment requires improved understanding of the thalamocortical circuits and cellular mechanisms that support seizure initiation, propagation, and termination. With its dense, nucleus-specific reciprocal connectivity with cortical and limbic regions and intrinsic oscillatory properties, the thalamus is well-constructed and well-positioned to influence epileptic activity. Importantly, distinct thalamic nuclei exhibit differing engagement depending on seizure type and propagation patterns. Here, we review current knowledge of thalamocortical anatomy and function, cellular mechanisms of ictal propagation, and the role of the thalamus in generalized and focal seizures, with emphasis on human studies. We further examine how these anatomical and mechanistic insights inform neuromodulatory interventions aimed at improving seizure control.

PMID:42456232 | DOI:10.1016/j.neurot.2026.e00967

J Forensic Leg Med. 2026 Jul 15;122:103215. doi: 10.1016/j.jflm.2026.103215. Online ahead of print.

ABSTRACT

OBJECTIVE: To assess whether transfer-learning models applied to panoramic radiographs (PANs) can classify individuals at the threshold of legal majority (≥18 years).

MATERIALS AND METHODS: Vision Transformer (ViT) and EfficientNetV2 models were trained on PANs from Bosnian and Lebanese individuals aged 14-24.99 years (n = 1764), considering pooled and sex-specific datasets with and without augmentation. Binary classification, multiclass classification, and regression models were trained and evaluated. Model performance on an internal test set derived from the same sample was summarized using accuracy, sensitivity, specificity, F1 score, and area under the receiver operating characteristic curve (ROC AUC). For regression and multiclass models, legal majority classification was additionally assessed by thresholding predicted ages or age categories at 18 years. External validation employed an independent Brazilian dataset (n = 1579; 14-24.99 years). Formal statistical comparison between internal and external performance employed two-proportion z-tests for accuracy and DeLong’s test for ROC AUC.

RESULTS: On the internal test set, the best-performing binary classification model, EfficientNetV2 with augmentation on the pooled dataset, achieved an accuracy of 0.90, sensitivity of 0.92, specificity of 0.86, F1 score of 0.91, and ROC AUC of 0.93. Using the same pooled, augmented configuration, thresholded regression predictions achieved accuracies of 0.85 for EfficientNetV2 and 0.83 for ViT, whereas thresholded multiclass predictions achieved accuracies of 0.84 and 0.73, respectively. Compared with direct binary classification, these thresholded outputs showed no clear advantage, and multiclass models generally showed higher specificity but lower sensitivity. This same pattern was retained when thresholded models were evaluated on the external validation set, with thresholded regression remaining comparatively stable and thresholded multiclass performance declining more markedly, especially for ViT. Visualization maps (gradient-weighted class activation mapping [Grad-CAM] and occlusion sensitivity) confirmed attention to relevant dental structures. On external validation, the same binary EfficientNetV2 configuration achieved an accuracy of 0.81, sensitivity of 0.90, specificity of 0.61, F1 score of 0.87, and ROC AUC of 0.85. Sex-specific models performed similarly, showing no clear advantage over pooled training. Statistical testing confirmed significant AUC degradation on external validation for both models (EfficientNetV2: p = 0.004; ViT: p < 0.001), while the accuracy drop was significant only for ViT (p = 0.035).

CONCLUSIONS: This study demonstrated that transfer learning with EfficientNetV2 and Vision Transformer can distinguish minors from adults using PANs with high internal performance and acceptable external generalization. Direct binary classification provided the most robust approach for legal age assessment in the present dataset. The findings support the forensic potential of these models, while also indicating the need for further work to improve robustness and real-world applicability.

PMID:42456231 | DOI:10.1016/j.jflm.2026.103215

Nurse Educ Today. 2026 Jul 9;166:107272. doi: 10.1016/j.nedt.2026.107272. Online ahead of print.

ABSTRACT

BACKGROUND: With the rapid integration of artificial intelligence (AI) into the nursing field, AI literacy has emerged as a critical competency for nursing students. However, evidence regarding the current status and influencing factors of AI literacy among Chinese nursing students remains limited.

AIM: This cross-sectional study aimed to examine AI literacy levels across four dimensions (Awareness, Application, Evaluation, Ethics) among Chinese nursing students, identify its independent influencing factors, and generate localized empirical evidence for AI curriculum construction in vocational nursing education.

METHODS: A cross-sectional study was conducted between November 1, 2025 and January 20, 2026. The research institution has not set up an institutional ethics committee; the whole protocol was ethically reviewed by the supervisor team in line with the Declaration of Helsinki before recruitment. A total of 178 sophomore nursing students from Henan Vocational University of Science and Technology were enrolled via convenience sampling. Data were collected using the Artificial Intelligence Literacy Scale (AILS, Wang et al., 2023). This scale is publicly accessible for non-commercial academic research; the original authors have granted open-use permission for educational cross-sectional surveys without additional formal written authorization. After data cleaning, 155 valid questionnaires were included (effective response rate = 86.9%). Nonparametric tests and ordinal logistic regression were performed via IBM SPSS 27.0 for statistical analysis.

RESULTS: Participants demonstrated a moderate level of overall artificial intelligence literacy (Median = 60.00, IQR = 21.00, SD = 13.18). The Application dimension scored highest. Gender and interest in AI were independent influencing factors. Strong positive correlations were identified among all four dimensions.

CONCLUSIONS: Study findings suggest that nursing education should integrate AI content and targeted training to strengthen students’ AI awareness, critical evaluation, and ethical awareness, while providing tailored support for female students and boosting AI interest.

PMID:42456212 | DOI:10.1016/j.nedt.2026.107272

J Biomech. 2026 Jul 3;205:113451. doi: 10.1016/j.jbiomech.2026.113451. Online ahead of print.

ABSTRACT

Biomechanical models of myocardial growth and remodelling (G&R) are often modelled as responses to deviations from a preferred biomechanical homeostatic state, triggered by stretch or stress. Yet how these critical values of stretch and stress should be defined remains unresolved. This study examines how different homeostatic threshold definitions influence predicted left ventricular G&R. Myocardial growth was implemented using an updated-reference constrained mixture framework, with eccentric growth driven by diastolic myofibre stretch and concentric growth driven by systolic active stress. Homeostatic thresholds were derived from the baseline LV dynamics. Our results demonstrate that the choice of critical homeostatic thresholds plays an important role in triggering cardiac G&R. For volume overload, the mean and 50th percentile end-diastolic myofibre stretch show the potential to approach a preferred mechanical state with preserved cardiac pump function. For pressure overload, a 90th percentile threshold of peak active stress appears to normalise systolic stress to baseline levels with sufficient cardiac output. Localised thresholds generally lead to more heterogeneous growth patterns and potentially high stretch/stress concentrations. These findings highlight the importance of defining homeostatic targets as tissue-level quantities and provide practical guidance to improve cardiac G&R modelling.

PMID:42456208 | DOI:10.1016/j.jbiomech.2026.113451

J Pediatr Nurs. 2026 Jul 15;90:541-546. doi: 10.1016/j.pedn.2026.07.016. Online ahead of print.

ABSTRACT

PURPOSE: To investigate the current status of family management among children after enterostomy surgery and identify its influencing factors, providing an evidence-based foundation for developing practical resources and policy recommendations to reduce family caregiving burdens and improve postoperative rehabilitation outcomes.

METHODS: This cross-sectional study included 155 pediatric enterostomy patients and their family caregivers from June 2025 to March 2026. Family management and family function were assessed using the Chinese versions of the Family Management Measure (FaMM) and the Family Assessment Device-General Functioning (FAD-GF). We analyzed the clinical data of patients and caregivers, identified factors influencing family management, and conducted statistical analyses on these factors.

RESULTS: The average FaMM score was 162.11 ± 11.34, indicating a moderate level. Univariate analysis showed that whether the child was an only child, child’s age, family’s place of residence, disease course, caregiver education level, employment status, and average family monthly income were all influencing factors (P < 0.05). Pearson correlation analysis indicated a positive correlation between family management and family function (r = 0.934, P < 0.001). Multiple linear regression analysis revealed that family function, disease course, caregiver employment status, and average family monthly income were the primary influencing factors of family management, explaining 77.6% of the total variance.

CONCLUSION: The level of family management among children with enterostomy needs improvement. Better family function, longer disease course, employed caregivers, and higher family income are associated with higher levels of family management. Clinically, it is essential to provide these families with professional information and policy support to enhance their family management capabilities.

PMID:42456196 | DOI:10.1016/j.pedn.2026.07.016

J Am Chem Soc. 2026 Jul 15. doi: 10.1021/jacs.6c07565. Online ahead of print.

ABSTRACT

Mixing wheat flour with water transforms an otherwise dispersible powder into a cohesive and extensible dough capable of sustaining large deformations. The emergence of this malleable material is due to polymerization of glutenin subunits. However, the molecular principles governing this transformation remain unclear. Here, we investigate the hydration-induced association behavior of the N-terminal domain of the high-molecular-weight glutenin subunit 1Dx5 (1Dx5-NTD). Upon hydration, 1Dx5-NTD adopts a collapsed, heterogeneous ensemble rather than a single, well-defined folded structure and undergoes intermolecular association into assemblies ranging from trimers to ∼70,000-mers. These associations are primarily determined by favorable enthalpic interactions that are modulated by moderate ionic strength. Hydrophobic residues (Tyr52, Phe65, Tyr66, and Trp82) are buried within the collapsed monomer yet indirectly modulate protein association through correlated ensemble dynamics, as supported by mutational analysis. Importantly, 1Dx5-NTD contains three cysteine residues (Cys10, Cys25, and Cys40) that undergo disulfide bond formation during association. While multiple disulfide linkage patterns are chemically possible, mass spectrometry-based cross-linking analysis reveals a limited subset of preferential disulfide linkages. This finding suggests that disulfide formation is biased by the conformational ensemble and local cysteine accessibility, rather than occurring randomly. Together, these results support a model in which glutenin polymerization emerges from ensemble-averaged intermolecular contacts coupled with statistically biased disulfide bond formation, giving rise to a cohesive, multidimensional protein network without requiring a uniquely defined folded structure. This work highlights how collapsed protein ensembles can encode chemical selectivity and macroscopic material properties.

PMID:42456183 | DOI:10.1021/jacs.6c07565

JMIR Res Protoc. 2026 Jul 15;15:e92270. doi: 10.2196/92270.

ABSTRACT

BACKGROUND: Posttraumatic stress disorder (PTSD) affects up to 25% of trauma-exposed adolescents; yet, access to evidence-based treatment remains limited in rural regions. Trauma-focused cognitive behavioral therapy (TF-CBT) is the first-line intervention, but structural barriers such as long travel distances and therapist shortages hinder implementation. Digital and blended formats may improve accessibility, but evidence for adolescents with PTSD is limited. Shame and self-criticism are common following interpersonal trauma and can reduce engagement; compassion-focused strategies target these mechanisms and aim to enhance emotional safety.

OBJECTIVE: This protocol outlines a pilot randomized controlled trial (RCT) evaluating the feasibility of delivering blended trauma-focused cognitive behavioral therapy with compassion (bTF-CBT-C) for adolescents with PTSD in routine clinical care. A secondary aim is to assess the acceptability of the intervention among adolescents and caregivers. In addition, the study explores patterns and variability in clinical outcomes to inform the design of a future noninferiority trial.

METHODS: A 2-arm parallel-group pilot RCT will randomize 40 adolescents (12-17 years) with DSM-5 (Diagnostic and Statistical Manual of Mental Disorders [Fifth Edition]) PTSD to bTF-CBT-C or standard TF-CBT in routine child and adolescent psychiatric services. The intervention includes an approximately 5-week web-based stabilization phase, followed by 7-14 therapist-led sessions delivered primarily via videoconference, with some in-person sessions. Primary outcomes will assess feasibility (recruitment, retention, adherence, data completeness, and adverse events) and acceptability (satisfaction, alliance, and qualitative interviews). Exploratory outcomes include PTSD symptoms, self-compassion, emotion regulation, depression, anxiety, suicidality, and dissociation. Assessments will be conducted at baseline, poststabilization, posttreatment, and 6-month follow-up. Feasibility and acceptability will be summarized descriptively. Exploratory analyses using analysis of covariance and mixed effects models will estimate variance parameters, confidence intervals, and descriptive change trajectories without hypothesis testing. Qualitative data will be analyzed using reflexive thematic analysis.

RESULTS: Recruitment started in February 2026, and data collection is projected to be completed by December 2028. Feasibility and acceptability outcomes, along with exploratory clinical patterns, will be reported in accordance with the CONSORT (Consolidated Standards of Reporting Trials) extension for pilot and feasibility trials.

CONCLUSIONS: Findings will inform the refinement of the intervention, the assessment of trial feasibility, and the selection of outcomes for a fully powered noninferiority RCT. The study will also contribute to understanding how compassion-focused strategies may support emotional safety and engagement in trauma-focused treatment for adolescents.

PMID:42456165 | DOI:10.2196/92270

JMIR Form Res. 2026 Jul 15;10:e89719. doi: 10.2196/89719.

ABSTRACT

BACKGROUND: The postpartum period represents a critical window for maternal health, yet many individuals lack sustained support and timely identification of physical and mental health risks. Digital health interventions offer a scalable approach to extend care beyond clinical settings. Yet, key elements, including real-world challenges, usability, and effectiveness of such platforms, are insufficiently characterized in this literature.

OBJECTIVE: This study aimed to conduct a formative randomized evaluation to assess the feasibility, engagement, and preliminary signals of impact of the Joyuus platform and to examine the potential for early identification of postpartum health risks.

METHODS: We conducted a 12-week randomized evaluation with postpartum participants recruited through community-based organizations. Participants were randomized to either the Joyuus intervention or standard postpartum care. Primary and secondary outcomes included the Barkin Index of Maternal Functioning, the Edinburgh Postnatal Depression Scale (EPDS), the State-Trait Anxiety Inventory, and the Connor-Davidson Resilience Scale. Analyses were conducted using an intention-to-treat approach with linear regression models adjusting for baseline values. Engagement metrics (ie, sessions, time on site, and feature use) were captured through in-app analytics. An exploratory predictive model was developed using baseline clinical, behavioral, and demographic variables to identify individuals at risk for postpartum depression.

RESULTS: Baseline characteristics were generally balanced across arms, and no statistically significant differences were observed in education, income, or marital status. No statistically significant differences were found between treatment and control groups in the 12-week changes for the primary or secondary outcomes. Mean EPDS scores were 9.7 in the intervention group and 9.3 in the control group. In the sample, 60 (45.5%) of the 132 participants met the criteria for elevated depression (EPDS score ≥11 or a positive response to question 10 on self-harm), indicating a high burden of symptoms within the study population. Engagement with the platform was highest during the first 4 weeks. Among intervention participants, 80% created an account. Participants reported high levels of perceived usefulness, ease of use, and relevance of content. Qualitative analysis of open-ended survey responses highlighted limited awareness of postpartum-specific resources and a preference for simple, accessible information. An exploratory predictive model demonstrated a recall of 0.89 and a precision of 0.73 in identifying individuals at risk for postpartum depression, suggesting the feasibility of early risk identification using integrated data inputs.

CONCLUSIONS: Joyuus demonstrated feasibility and acceptability but did not produce statistically significant improvements in maternal functioning or maternal health outcomes over 12 weeks. Joyuus identified high rates of depressive symptoms and early engagement patterns, which suggest an opportunity for earlier identification of risk and intervention during the postpartum period. Exploratory modeling results indicate the potential for data-driven approaches to support earlier detection. Future work includes optimizing engagement strategies and expanding validation of predictive detection to improve postpartum surveillance and outcomes.

PMID:42456164 | DOI:10.2196/89719

JMIR Med Educ. 2026 Jul 15;12:e77034. doi: 10.2196/77034.

ABSTRACT

BACKGROUND: The use of remote teaching in medical education has increased since the COVID-19 pandemic. However, the effectiveness of synchronous remote teaching for specific psychomotor components of the neurological examination, such as tendon reflex assessment, remains underexplored.

OBJECTIVE: This study aimed to evaluate the short-term effectiveness of synchronous remote teaching compared with traditional in-person teaching for training medical students in tendon reflex examination skills and related neurological examination knowledge.

METHODS: This quasi-experimental study enrolled 110 fifth-year medical students between August 2022 and April 2024, who were assigned to either synchronous remote teaching (n=46) or in-person teaching (n=64). The participants completed pre- and postcourse knowledge tests and an objective structured clinical examination (OSCE) assessing tendon reflex examination skills. Motivational beliefs related to self-regulated learning and technology acceptance were measured at baseline. Group differences were analyzed using t tests, Mann-Whitney U tests, and correlational analyses.

RESULTS: Posttest knowledge scores were higher in the synchronous remote teaching group than in the in-person group (mean difference 0.86, 95% CI 0.21-1.51; Cohen d=0.50; P=.01), indicating a modest effect size. Both groups demonstrated substantial improvements in tendon reflex OSCE performance; however, the between-group difference in OSCE score improvement was not statistically significant (mean difference -3.64; 95% CI -9.36 to 2.08; Cohen d=-0.26; P=.21). Within the synchronous remote teaching group, higher motivational beliefs were positively correlated with OSCE score gains (Pearson r=0.34; P=.04; n=46), suggesting a potential role of self-regulated learning-related motivational processes. These exploratory findings should be interpreted with caution.

CONCLUSIONS: Under the conditions of this study, synchronous remote teaching was associated with modest improvements in knowledge outcomes and with OSCE-assessed tendon reflex examination skill gains that were comparable to those of traditional in-person teaching, without statistically significant between-group differences in psychomotor performance. These findings do not establish durable skill acquisition and should not be interpreted as evidence of equivalence or superiority of synchronous remote teaching over in-person instruction. Unlike many prior remote medical education studies that primarily focused on asynchronous learning or self-reported outcomes, this study incorporated synchronous interactive instruction with objective OSCE-based psychomotor assessment. These findings contribute to the emerging literature on synchronous remote clinical skills education, although future randomized and longitudinal studies are needed to evaluate long-term skill retention and broader applicability.

PMID:42456133 | DOI:10.2196/77034