Nevin Manimala

Brain Inform. 2026 Apr 3. doi: 10.1186/s40708-026-00299-w. Online ahead of print.

ABSTRACT

PURPOSE: MELD Graph is a state-of-the-art artificial intelligence (AI) model for automated detection of focal cortical dysplasia (FCD), but its performance remains limited, highlighting the need to investigate which aspects of the pipeline affect its accuracy.

METHODS: A retrospective failure-mode analysis of the MELD Graph pipeline was performed in 242 subjects, with model predictions and FreeSurfer segmentations reviewed to classify errors as segmentation-associated or algorithm-related. FCD imaging features salient to humans were quantified, with statistical associations examined for both MELD Graph detection and focal FreeSurfer segmentation failure.

RESULTS: MELD Graph demonstrated overall performance similar to previously published non-harmonized results, achieving a sensitivity of 69%, specificity of 44%, and positive predictive value (PPV) of 75%. Focal FreeSurfer segmentation failures were associated with 21% of false negative patients, 25% of false positive clusters in patients, and 16% of false positive clusters in controls. Following manual cortical segmentation correction and rerunning of MELD Graph, 67% of the segmentation-associated missed lesions were detected, and segmentation-associated false positive clusters were reduced or eliminated in 75% of controls with such clusters. Higher conspicuity on T1-weighted images was associated with MELD Graph detection, whereas greater conspicuity on T2-FLAIR images relative to T1 was associated with detection failure. Non-bottom-of-sulcus lesion location, higher human conspicuity measures, and low T1 image quality were positively associated with focal FreeSurfer segmentation failures.

CONCLUSION: FreeSurfer segmentation failures are a significant potential source of error in the MELD Graph pipeline. FCD imaging features salient to humans and image quality were also associated with variability in algorithm performance. Robust cortical segmentation and stronger integration of T2-FLAIR imaging features may be beneficial for automated FCD detection tools.

CLINICAL TRIAL REGISTRATION: Not applicable. This study is a retrospective analysis of previously acquired open-source imaging datasets and does not constitute a clinical trial.

PMID:41931246 | DOI:10.1186/s40708-026-00299-w

Drugs R D. 2026 Apr 3. doi: 10.1007/s40268-026-00543-y. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVES: Viloxazine, which has been used to treat depression and attention-deficit/hyperactivity disorder (ADHD), has been termed a moderate-affinity, selective norepinephrine reuptake inhibitor based on high selectivity for norepinephrine relative to serotonin and dopamine transporters. However, accumulated research suggests a more complex mechanism of action, based on studies showing activity at serotonin 5-HT_2C, 5-HT_2B, and 5-HT₇ receptors, as well as findings that viloxazine increases extracellular serotonin (along with norepinephrine and dopamine) in the rat prefrontal cortex. This in vitro pharmacology study aimed to replicate and expand prior experiments to better characterize viloxazine’s affinity for and activity at the norepinephrine transporter (NET) and individual serotonin receptors and to clarify how these effects contribute to the mechanism of action.

METHODS: Using in vitro binding competition and functional assays and ex vivo receptor occupancy studies in rats, we assessed viloxazine activity at human NET isoforms and 5-HT_2C, 5-HT_2B, and 5-HT₇ receptors relative to clinically relevant unbound viloxazine plasma concentrations (0.4-3.6 μM).

RESULTS: Viloxazine showed moderate binding affinity for NET (inhibition constant [K_i] = 0.13 µM) and 5-HT_2C (K_i = 0.66 µM), 5-HT_2B (K_i = 0.83 µM), and 5-HT₇ (K_i = 1.90 µM) receptors. In vitro functional studies showed viloxazine acted as a NET inhibitor, 5-HT_2C partial agonist, and 5-HT_2B and 5-HT₇ antagonist. At clinically relevant concentrations, viloxazine could potentially occupy nearly 95% of NET, more than 80% of 5-HT_2C and 5-HT_2B, and 65% of 5-HT₇ receptors. Subsequent ex vivo studies in rats confirmed high NET occupancy (67-94%) at clinically relevant concentrations.

CONCLUSIONS: These results validate previous experiments showing that viloxazine, in addition to displaying high NET occupancy, acts as a partial agonist at 5-HT_2C receptors and an antagonist at 5-HT_2B and 5-HT₇ receptors at clinically relevant concentrations for ADHD treatment. Therefore, both NET inhibition and serotonin receptor activity may contribute to viloxazine’s clinical efficacy. These findings are contributing to a renewed understanding of viloxazine’s pharmacodynamic profile and likely multimodal mechanism of action.

PMID:41931242 | DOI:10.1007/s40268-026-00543-y

Trop Anim Health Prod. 2026 Apr 3;58(3):217. doi: 10.1007/s11250-026-05018-5.

NO ABSTRACT

PMID:41931203 | DOI:10.1007/s11250-026-05018-5

CNS Neurol Disord Drug Targets. 2026 Mar 26. doi: 10.2174/0118715273434957260202115619. Online ahead of print.

ABSTRACT

INTRODUCTION: Inflammatory and coagulation pathways are crucial in the pathogenesis and clinical progression of ischemic stroke. The objective of this study was to evaluate serum concentrations of interleukin-2 (IL-2) and interleukin-10 (IL-10), as well as the gene expression of tissue factor (TF) in peripheral blood mononuclear cells (PBMCs), and to examine their correlations with stroke severity and clinical outcomes.

MATERIALS AND METHODS: We enrolled 148 patients with ischemic stroke and 30 healthy controls matched for age and sex in a cross-sectional design. We used ELISA to measure the levels of IL-2 and IL-10 in serum and real-time PCR to look at TF gene expression in PBMCs. The NIH Stroke Scale (NIHSS) was used to measure the severity of strokes, and the results were compared to clinical variables.

RESULTS: Patients with severe stroke showed significantly lower levels of IL-2 and IL-10 (p < 0.001), and TF expression in PBMCs was significantly higher in both mild and severe stroke groups compared to controls (p < 0.001). There was no statistically significant difference between the mild and severe groups (p = 0.213). In severe cases, IL-2 and TF were negatively correlated (p = 0.036). Nonetheless, none of the biomarkers independently forecasted survival outcomes.

DISCUSSION: The results show that the immune-coagulation axis is not working properly in severe ischemic stroke. Lower levels of IL-2 and IL-10 may indicate that regulatory T-cells aren’t working properly and that anti-inflammatory control isn’t working, which can cause monocytes to become active and TF levels to rise. This interaction probably makes thromboinflammatory cascades worse, which leads to more damage to the nervous system. These changes, even though they don’t predict survival, give us a better understanding of how strokes work and open up new possibilities for targeted immunomodulatory therapy.

CONCLUSION: The changes in IL-2, IL-10, and TF expression indicate a coordinated disruption of immune and thrombotic pathways in individuals with severe ischemic stroke. Although not prognostic of mortality, these biomarkers may indicate disease severity and represent potential targets for future therapeutic interventions. Longitudinal studies are necessary to validate their prognostic significance and investigate their incorporation into clinical decision-making algorithms for stroke.

PMID:41930584 | DOI:10.2174/0118715273434957260202115619

Cancer Treat Res Commun. 2026 Mar 12;47:101171. doi: 10.1016/j.ctarc.2026.101171. Online ahead of print.

ABSTRACT

BACKGROUND: TFE3-RCC is rare, hard to distinguish from common RCC on CT, posing preoperative diagnostic challenges for clinicians. This two-center study aimed to develop interpretable machine learning models using radiomics to differentiate Xp11.2/TFE3 translocation renal cell carcinoma (TFE3-RCC) from common renal cell carcinoma (RCC) subtypes using computed tomography (CT) images.

METHODS: Retrospective data from 1394 patients (39 TFE3-RCC, 1355 non-TFE3 RCC) were analyzed. A propensity score matching (PSM) was applied, resulting in 234 cases (TFE3: n = 39, non-TFE3: n = 195) included in the radiomics study. CT images were segmented using an AI-based model, and 102 radiomic features (shape, first-order statistics, texture) were extracted. Recursive feature elimination (RFE) with random forest and gradient boosting models were used for feature selection and model development. Performance was evaluated via area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity.

RESULTS: The patients with TFE3-RCC were significantly younger (36.51 ± 12.68 vs. 57.30 ± 12.00 years, P < 0.05), and had more frequent calcification (30.8% vs. 6.4%, P < 0.05) and were larger (5.50 ± 3.17 cm vs. 4.11 ± 2.06 cm, P = 0.005) than those with non-TFE3 RCC, and preferred to implicate females (female: 46.2% vs. 29.3%, P = 0.023). The model identified six optimal features, with skewness (relative weight: 44.57%) and first-order statistics as key predictors. The training set and test set achieved stable performances with AUC (0.951 (95% CI: 0.920-0.983) and 0.864 (95% CI: 0.749-0.979)) and accuracy (0.878 and 0.852).

CONCLUSION: Interpretable radiomics-based machine learning models effectively differentiate TFE3-RCC from common RCC subtypes, with skewness and intensity features as critical biomarkers. This approach may improve preoperative diagnosis, though larger multi-center studies and integration of multi-omics data are needed for clinical translation.

PMID:41930555 | DOI:10.1016/j.ctarc.2026.101171

Int Emerg Nurs. 2026 Apr 1;86:101815. doi: 10.1016/j.ienj.2026.101815. Online ahead of print.

ABSTRACT

BACKGROUND: Workplace violence (WPV) is a major occupational hazard for emergency nurses, with verbal, psychological, and physical threats undermining safety. Multiple safety strategies have been proposed, yet little research has examined their prevalence or link to perceived safety.

METHODS: A cross-sectional design was employed using an anonymous electronic survey of emergency nurses across 17 U.S. states. The survey, validated by nursing experts, assessed demographics, workplace characteristics, exposure to WPV, and 14 safety precautions. Perceived safety was rated on a 10-point scale. Descriptive statistics, bivariate tests, and multivariate regression with bootstrapping were conducted.

RESULTS: Among 134 participants (M age = 42.8 years, 84.3% female), 48.1% reported experiencing WPV in the past month. Mean safety rating was 6.84 (SD = 2). De-escalation and security presence were most prevalent (90.3%), followed by controlled access (80.6%) and security cameras (77.6%). Regression analysis showed urban nurses reported lower safety than suburban nurses (b = – 1.62, 95% CI [-2.438, -0.913], p < 0.001). Security presence, controlled access, and lighting were associated with higher safety perceptions (b = 1.31, 95% CI [0.487, 2.322], p = 0.018; b = 1.12, 95% CI [0.396, 1.881], p = 0.008; b = 1.09, 95% CI [0.478, 1.784], p = 0.004, respectively).

CONCLUSION: Findings highlight the prevalence of WPV in EDs and identify key safety interventions linked to nurses’ perceptions of safety. Results can inform policy and guide workplace improvements.

PMID:41930553 | DOI:10.1016/j.ienj.2026.101815

Neural Netw. 2026 Mar 22;201:108889. doi: 10.1016/j.neunet.2026.108889. Online ahead of print.

ABSTRACT

Federated learning, a vital paradigm in modern machine learning, enables private and decentralised training of models that is crucial for learning from sensitive data. Noisy label learning, another vital paradigm in modern machine learning, addresses the training of models from the data with potentially incorrect labels. Their integration, namely federated learning with noisy labels (FLNL), is an emerging but challenging topic arising from the practice of machine learning, which, however, still lacks a review of its research progress. The aim of this paper is to fill in this gap. We first summarise four core challenges to FLNL: localised label noise, across-client heterogeneity of label noise, localised overfitting to label noise, and inadequate benchmarking. We then propose a taxonomy to categorise current FLNL studies into four types that address the four challenges correspondingly: sample-wise methods, client-wise methods, model-wise methods, and benchmark-wise studies. This work offers the first comprehensive and concise review dedicated to FLNL; moreover, we also provide future research directions for this rapidly evolving and practically significant field.

PMID:41930546 | DOI:10.1016/j.neunet.2026.108889

J Pediatr Nurs. 2026 Apr 1;88:635-644. doi: 10.1016/j.pedn.2026.03.041. Online ahead of print.

ABSTRACT

BACKGROUND: Children with disabilities increasingly encounter digital environments with opportunities and risks, while relationships among digital parenting self-efficacy, parental self-efficacy, and family support remain understudied.

METHODS: This cross-sectional, descriptive, and correlational study was conducted with 195 parents of children aged 0-18 years with disabilities registered in a disability services unit in Türkiye. Data were collected using a sociodemographic form, the Digital Parenting Self-Efficacy Scale, the Parental Self-Efficacy Scale, and the Family Support Scale. Data were analyzed using descriptive statistics, Pearson correlation analysis, and multiple linear regression. Missing data were handled using multiple imputation with the fully conditional specification method, and pooled estimates were calculated according to Rubin’s rules.

RESULTS: Most children used the internet daily and for extended periods, and some exhibited behaviors suggestive of problematic use. Parents demonstrated moderate-to-high levels of digital literacy, digital security awareness, and digital communication skills. Parental self-efficacy showed positive associations with digital competencies and perceived family support. In the adjusted regression model controlling for maternal age, maternal education, family income, and social security status, digital security and perceived family support emerged as significant predictors of parental self-efficacy, with digital security representing the strongest predictor.

CONCLUSIONS: Digital security and family support play important roles in strengthening parental self-efficacy among families raising children with disabilities.

IMPLICATIONS FOR PEDIATRIC NURSING: Pediatric nurses can play a key role in assessing digital use in families of children with disabilities, strengthening parents’ digital literacy and self-efficacy, and designing family-centred interventions to reduce digital risks and enhance family support.

PMID:41930534 | DOI:10.1016/j.pedn.2026.03.041

Heart Lung. 2026 Apr 1;78:102775. doi: 10.1016/j.hrtlng.2026.102775. Online ahead of print.

ABSTRACT

BACKGROUND: Inspiratory muscle weakness is common among patients with chronic heart failure undergoing cardiac resynchronization therapy (CRT). Home-based inspiratory muscle training (IMT) could be a vital therapeutic strategy, particularly for those with limited access to cardiac rehabilitation programs.

OBJECTIVES: This study aimed to investigate the effects of a 12-week home-based IMT program on inspiratory muscle strength, cardiac function, NT-proBNP levels, exercise capacity, quality of life, and daily activities. It examined the relationships between changes in NT-proBNP levels, walking distance, maximal inspiratory pressure (MIP), and left ventricular ejection fraction (LVEF) in patients with heart failure at least one year after CRT implantation.

METHODS: In this randomized controlled trial, 32 patients with CRT devices were assigned to either the IMT group (n = 19) or the control group (n = 13). Outcome measures included MIP, LVEF, NT-proBNP levels, the 6-minute walk test, performance of activities of daily living (PMADL-8), and quality of life (Nottingham Health Profile-NHP).

RESULTS: Compared to the control group, the IMT group showed statistically significant improvements in MIP (p < 0.001), LVEF (p = 0.025), NT-proBNP levels (p = 0.003), walk distance (p < 0.001), PMADL-8 score (p < 0.001), and NHP total score. Significant correlations were observed among the changes in NT-proBNP, MIP, LVEF, and walk distance.

CONCLUSION: In this small-sample study home-based IMT was associated with improvements in respiratory strength, cardiac function, functional capacity, patient-reported functionality, and quality of life. This confirms the role of home-based IMT as a supportive preventive strategy for CRT patients who lack access to rehabilitation. Larger-scale, long-term studies are needed to confirm its effects on remodeling and clinical outcomes.

PMID:41930533 | DOI:10.1016/j.hrtlng.2026.102775

Comput Biol Chem. 2026 Mar 27;123:109039. doi: 10.1016/j.compbiolchem.2026.109039. Online ahead of print.

ABSTRACT

Disruption of transcription termination (DoTT) occurs when RNA polymerase II reads past a gene’s normal 3′ end, generating downstream “readthrough” RNA. DoTT has been reported under stresses such as viral infection and metabolic perturbation. But, many existing detection tools analyze samples one at a time or rely on rigid downstream windows, limiting direct condition-to-condition testing. We present DoTT-ML, a condition-aware pipeline for detecting transcription termination disruption from conventional short-read RNA-seq. This pipeline extends gene annotations downstream by a tunable window, applies an optional gap to reduce termination-proximal noise, and applies differential analysis between conditions using a robust statistical workflow. An optional machine learning approach provides a post-hoc prioritization when curated reference annotations are available. We benchmarked DoTT-ML against ARTDeco and DoGFinder across three public datasets: influenza A virus total RNA-seq, HSV-1 nascent 4sU-RNA, and HSV-1 Z-RNA RIP-seq. DoTT-ML showed comparably to, or better than, existing tools (high ROC AUC across datasets). Finally, in an in-house mouse, high-carbohydrate diet (HCD) liver model, DoTT-ML identified diet-associated readthroughs at metabolic genes. Experimental validation confirmed a stable readthrough transcript at the Scd1 locus under dietary stress, serving as a proof of principle for the pipeline’s biological relevance. Together, DoTT-ML provides a practical framework for condition-aware, readthrough detection and comparison across diverse RNA-seq assays.

PMID:41930502 | DOI:10.1016/j.compbiolchem.2026.109039