Category: Statistics

Brain Dev. 2025 Nov 15;47(6):104485. doi: 10.1016/j.braindev.2025.104485. Online ahead of print.

ABSTRACT

INTRODUCTION: Epilepsy is a chronic brain disorder characterized by recurrent seizures. Limited data on childhood epilepsy in Cameroon prompted this study.

METHODS: We conducted a cross-sectional study with retrospective data collection over six months (December 2023-May 2024). Medical records of children aged 3 months to 15 years, diagnosed with epilepsy and followed at Douala General Hospital between January 2020 and December 2023, were analyzed. Statistical analysis used SPSS 26.0, with Fisher’s exact and chi-square tests for associations, and logistic regression for predictive factors (p < 0.05).

RESULTS: 142 patients were included (male-to-female ratio = 1.21). Epilepsy prevalence was 2.4 %. Generalized seizures predominated (65.7 %) with focal epileptic abnormalities in 50.7 % of cases. Idiopathic generalized epilepsy represented 57.7 % of cases. Sodium valproate was used in 52.8 % of cases. Main etiological factors included: neonatal convulsions (61; 43 %), febrile seizures (49; 34.5 %) and neonatal asphyxia (35; 24.6 %). Seizures persisted in 35 patients (24.6 %) under treatment. Predictive factors for poor seizure control included unknown seizure type (OR 14.25 [1.10-183.97]; p = 0.04), cryptogenic focal epilepsy (OR 12.55 [1.58-99.71]; p = 0.02), and use of prayers or traditional medicines (OR 7.45 [1.01-55.13]; p = 0.05). Memory disorders significantly impacted school performance (OR 4.95 [1.80-13.59]; p = 0.002), along with lack of concentration (OR 3.04 [1.04-8.84]; p = 0.04).

CONCLUSION: This study identified specific predictive factors for poor epileptic control and confirms cognitive impact on schooling, providing intervention targets to optimize neurological and educational management in Cameroon.

PMID:41242021 | DOI:10.1016/j.braindev.2025.104485

Mar Pollut Bull. 2025 Nov 15;223:118998. doi: 10.1016/j.marpolbul.2025.118998. Online ahead of print.

ABSTRACT

Brown and red macroalgae are promising biomonitors of inorganic pollutants in marine environments, due to quick growth, high biomass, ease of sampling and the production of a diverse array of chelating agents. The limited distribution and low population densities of most of the currently adopted species, however, prompt the need to search for suitable alternatives. To this end, the accumulation kinetics of ten metals (Cd, Co, Cr, Cu, Fe, Mn, Ni, Pb, V, Zn) and one metalloid (As) under field-conditions were studied in Dictyota spiralis and Laurencia microcladia, brown and red macroalgae, respectively. Specifically, element concentrations were measured at 0, 2, 4, 8, 16 and 24d on algae (both alive and devitalized) exposed in the field (through purposely developed bags) in 4 sites differing in anthropogenic impacts along the Tyrrhenian coast (southern Italy). Robust Bayesian analyses were adopted for model fitting and selection, as well as to derive posterior distributions for parameters with straightforward practical implications for biomonitoring of coastal waters. Kinetics followed pseudo-first order, pseudo-second order and two-phase intraparticle diffusion models, with saturation times varying among elements, between species and between devitalization treatments. The spatial discrimination capability of the species varies in relation to the element, with a distinct advantage of living L. microcladia for the biomonitoring of Co, Cr, Cu, Fe, Mn and V, and of devitalized D. spiralis for the biomonitoring of As, Cu, Ni, Pb and Zn. Overall, findings highlight the remarkable effectiveness of transplants of both the species for the active biomonitoring of marine coastal ecosystems, especially when leveraging over their complementary selectivity toward different elements. The specificity in accumulation behavior allows also diversifying the target use of the two species, suggesting shorter-term monitoring and even bioremediation applications in the case of D. spiralis.

PMID:41242013 | DOI:10.1016/j.marpolbul.2025.118998

Comput Med Imaging Graph. 2025 Nov 14;126:102667. doi: 10.1016/j.compmedimag.2025.102667. Online ahead of print.

ABSTRACT

BACKGROUND AND OBJECTIVES: Neurofibromatosis type 1 (NF1) is a genetic disorder characterized by the development of multiple neurofibromas (NFs) throughout the body. Accurate segmentation of these tumors in whole-body magnetic resonance imaging (WB-MRI) is critical for quantifying tumor burden and clinical decision-making. This study aims to develop a pipeline for NF segmentation in fat-suppressed T2-weighted WB-MRI that incorporates anatomical context and radiomics to improve accuracy and specificity.

METHODS: The proposed pipeline consists of three stages: (1) anatomy segmentation using MRSegmentator and refinement with a high-risk NF zone; (2) NF segmentation using an ensemble of 3D anisotropic anatomy-informed U-Nets; and (3) tumor candidate classification using radiomic features to filter false positives. The study used 109 WB-MRI scans from 74 NF1 patients, divided into training and three test sets representing in-domain (3T), domain-shifted (1.5T), and low tumor burden scenarios. Evaluation metrics included per-scan and per-tumor Dice Similarity Coefficient (DSC), Volume Overlap Error (VOE), Absolute Relative Volume Difference (ARVD), and per-scan F1 score. Statistical significance was assessed using Wilcoxon signed-rank tests with Bonferroni correction.

RESULTS: On the in-domain test set, the proposed ensemble of 3D anisotropic anatomy-informed U-Nets with tumor candidate classification achieved a per-scan DSC of 0.64, outperforming 2D nnU-Net (DSC: 0.52) and 3D full-resolution nnU-Net (DSC: 0.54). Performance was maintained on the domain-shift test set (DSC: 0.51) but declined on low tumor burden cases (DSC: 0.23). Preliminary inter-reader variability analysis showed model-to-expert agreement (DSC: 0.67-0.69) comparable to inter-expert agreement (DSC: 0.69).

CONCLUSIONS: The proposed pipeline achieves the highest performance among established methods for automated NF segmentation in WB-MRI and approaches expert-level consistency. The integration of anatomical context and radiomics enhances robustness. Nonetheless, segmentation performance decreases in low tumor burden scenarios, indicating a key area for future methodological improvements. Additionally, the limited inter-reader agreement observed among experts underscores the inherent complexity and ambiguity of the NF segmentation task.

PMID:41241992 | DOI:10.1016/j.compmedimag.2025.102667

Phys Med. 2025 Nov 15;140:105215. doi: 10.1016/j.ejmp.2025.105215. Online ahead of print.

ABSTRACT

PURPOSE: Converting radiotherapy (RT) data from DICOM-RT into datasets suitable for statistical modelling remains challenging. We developed the DICOM Extraction and Structuration Toolkit (DEST), an automated solution that streamlines data extraction and ensures compatibility with statistical software. The reliability of DEST was also assessed by comparing its outputs with those from treatment planning systems (TPS) in study of cardiopulmonary dose-volume histograms (DVH) after radiotherapy (RT) for localised breast cancer.

METHODS: DEST comprises two main modules: a data extraction module and a viewer/analysis module. It processes DICOM-RT objects, including RT structure sets, RT plans, and RT dose files. Extractions are performed per treatment for predefined patient lists, after which structured data is consolidated. A 3D visualisation module verifies dose distributions for selected regions of interest, ensuring consistency and accuracy.

RESULTS: DEST was successfully applied to 404 patients from the “CANcer TOxicities – Radiation Therapy” (CANTO-RT) cohort. In this initial implementation, DEST showed strong overall agreement with regard to TPS for heart and lung dose metrics, including mean doses and dose-volume measurements. Specifically, near-minimum dose, median dose, near-maximum dose and percentages of volume receiving at least 10 Gy (V_10Gy), 20 Gy (V_20Gy), 30 Gy (V_30Gy) and 40 Gy (V_40Gy) showed high consistency between DEST and TPS.

CONCLUSIONS: DEST enhances accessibility to dose-volume metrics and will facilitate advanced modelling of medical outcomes (efficiency and risk) at the voxel level. By providing streamlined access to voxel spatial coordinates and local dose information, DEST enables more sophisticated analyses, such as clustering and localized region selection, supporting deeper insights into dose-response relationships.

PMID:41241989 | DOI:10.1016/j.ejmp.2025.105215

Clinics (Sao Paulo). 2025 Nov 15;80:100827. doi: 10.1016/j.clinsp.2025.100827. Online ahead of print.

ABSTRACT

BACKGROUND: While omentectomy is included in the guidelines for the surgical management of Borderline Ovarian Tumors (BOTs), it is unclear whether removal of a normal-appearing omentum confers a therapeutic advantage.

METHODS: The retrospective cohort study of patients with BOTs evalua0 ted the role of routine omentectomy and was followed by a meta-analysis to enhance the robustness of the findings. Data were obtained from patients treated at three Brazilian reference centers between January 2009 and October 2023. Progression-Free Survival (PFS), risk of death, and recurrence were compared between patients who underwent omentectomy and those who did not.

RESULTS: A total of 218 patients with BOTs were assessed: omentectomy was performed in 161 (73.8 %) and not performed in 57 (26.1 %). OS at 60 months was 95.5 % in the omentectomy group and 94.6 % in the non-omentectomy group (HR = 0.97 [95 % CI 0.20‒4.68]; p = 0.96). PFS was 97.2 % and 89.3 %, respectively (HR = 0.42; 95 % CI 0.10‒1.76; p = 0.23). Twelve studies comprising 2996 women with BOT, were included in the systematic review to evaluate the outcomes with and without omentectomy. Relative Risk (RR) of recurrence was 0.94 (95 % CI 0.67‒1.31; p = 0.7) for the non-omentectomy group compared with the omentectomy group. No statistically significant difference was observed, with an RR of 1.98 (95 % CI 0.24‒16.43; p = 0.53) for risk of death and an HR of 1.02 (95 % CI 0.25‒4.15; p = 0.98) for PFS.

CONCLUSION: The retrospective cohort study and meta-analysis showed a low incidence of metastatic disease in the omentum. No effect of omentectomy on OS, PFS, and recurrence in patients with BOT.

PMID:41241988 | DOI:10.1016/j.clinsp.2025.100827

Artif Intell Med. 2025 Nov 13;171:103305. doi: 10.1016/j.artmed.2025.103305. Online ahead of print.

ABSTRACT

Depression is a significant global health issue with increasing prevalence. Current diagnostic methods rely on subjective observations and questionnaires, often resulting in underestimation of the condition and insufficient treatment. This study investigates voice-based markers for detecting depressive symptoms through a novel system of virtual humans (VHs) capable of engaging in open-ended talks, unlike previous research which relied primarily on structured clinical interview formats. A total of 101 participants (42 with depressive symptoms) engaged in six casual social interactions with VHs simulating basic emotions, forming the DEPTALK dataset. Speech recordings and their automatic transcriptions were processed using state-of-the-art pre-trained transformer-based models to generate embeddings. We first employed a conversation-level aggregation strategy, combining embeddings across each dialogue and classifying them with Extreme Gradient Boosting. A single model trained on all six conversations per participant outperformed emotion-specific models, achieving F1 scores of 0.566 for speech, 0.329 for text, and 0.648 for the multimodal fusion, indicating that aggregating emotionally diverse interactions exposes stronger depression cues. To capture temporal dynamics, we further implemented a turn-level aggregation strategy using Gated Recurrent Units and training on all conversations. This approach improved performance for text (F1 = 0.505) and maintained competitive results for speech (F1 = 0.541), although the multimodal GRU model (F1 = 0.556) did not surpass the best conversation-level model. Overall, findings suggest that in casual conversations, depressive symptoms are primarily conveyed through prosody, with the addition of semantic context further enhancing detection. This study advances the understanding of speech-based depression patterns in simulated social interactions and highlights the potential of using VHs for more objective depressive symptoms detection.

PMID:41241976 | DOI:10.1016/j.artmed.2025.103305

Environ Technol. 2025 Nov 16:1-11. doi: 10.1080/09593330.2025.2587898. Online ahead of print.

ABSTRACT

New strategies for effluent treatment aimed at reducing environmental pollutants have significantly advanced, particularly biological methods involving enzymatic processes. In this context, this study evaluated the efficacy of a laccase-enriched enzymatic extract (specific laccase activity = 0.45 U/mg), obtained from the fungus Xylaria sp. for treating pharmaceutical effluents containing paracetamol, diclofenac, mefenamic acid, ibuprofen, and sulfamethoxazole, each at concentrations of 50 ppm. The enzymatic treatment resulted in notably higher degradation efficiencies for paracetamol and mefenamic acid under initial screening (∼70%). These drugs were selected for optimization due to their higher susceptibility to enzymatic degradation and because they are widely consumed pharmaceuticals frequently detected in aquatic environments. Afterward, optimization studies focused on these two pharmaceuticals, employing a statistical experimental design to determine optimal conditions, identified as pH 6.7, temperature of 40°C, and exposure time of 4.5 h. Under these optimized conditions, experimental results indicated a 95.55% reduction in paracetamol and a 55% reduction in mefenamic acid concentrations.Furthermore, enzyme immobilization on chitosan significantly enhanced stability and performance, maintaining approximately 90% reduction of both pharmaceuticals after multiple treatment cycles. These findings highlight the effectiveness of immobilized laccase systems and optimized reaction parameters, supporting their potential application for sustainable and efficient treatment of pharmaceutical effluent. Importantly, this work represents the first demonstration of using Xylaria sp. as a laccase source for pharmaceutical degradation, underlining its novelty and potential.

PMID:41241962 | DOI:10.1080/09593330.2025.2587898

Psychol Health Med. 2025 Nov 16:1-16. doi: 10.1080/13548506.2025.2584368. Online ahead of print.

ABSTRACT

This study investigated fertility-related quality of life among married women of reproductive age (20-45 years) with endometriosis-associated infertility, focusing on the mediating role of family resilience between infertility-related stress and fertility-related quality of life. A cross-sectional survey was conducted between January 2024 and April 2025 at a tertiary hospital gynecology and reproductive center in Guangdong Province, China. Data were collected via structured questionnaires and analyzed using IBM SPSS Statistics 25.0 and AMOS 26.0. Results indicated moderate levels of fertility-related stress (M = 70.92, SD = 13.98). Pearson correlation analysis revealed a significant negative association between fertility-related stress and fertility-related quality of life (r = -0.435, p < 0.01), while family resilience showed a positive correlation with fertility-related quality of life (r = 0.377, p < 0.01). Mediation analysis demonstrated that family resilience partially mitigated the negative impact of stress on fertility-related quality of life (standardized indirect effect = -0.11, 95% CI [-0.18, -0.06], p < 0.001). The findings suggest that while fertility-related stress directly impairs fertility-related quality of life in this population, family resilience serves as a protective buffer. Clinically, interventions aimed at strengthening family support and adaptive capacity may help reduce stress and improve well-being in women with endometriosis-associated infertility.

PMID:41241960 | DOI:10.1080/13548506.2025.2584368

STAR Protoc. 2025 Nov 14;6(4):104198. doi: 10.1016/j.xpro.2025.104198. Online ahead of print.

ABSTRACT

High-dimensional traits, like blood cell traits, are often analyzed using univariate genetic analysis approaches, ignoring trait relationships. Here, we present a protocol for using the flashfmZero software for analyses of latent factors that capture variation in observed traits generated by shared underlying biological mechanisms. We describe steps for calculating genome-wide association study (GWAS) summary statistics of latent factors from GWAS of observed traits, allowing for missing trait measurements. We then describe steps for jointly fine-mapping associations from multiple latent factors. For complete details on the use and execution of this protocol, please refer to Zhou et al.¹.

PMID:41241937 | DOI:10.1016/j.xpro.2025.104198

Biostatistics. 2024 Dec 31;26(1):kxaf037. doi: 10.1093/biostatistics/kxaf037.

ABSTRACT

Genome-wide association studies (GWASs) have highlighted the importance of pleiotropy in human diseases, where one gene can impact 2 or more unrelated traits. Examining shared genetic risk factors across multiple diseases can enhance our understanding of these conditions by pinpointing new genes and biological pathways involved. Furthermore, with an increasing wealth of GWAS summary statistics available to the scientific community, leveraging these findings across multiple phenotypes could unveil novel pleiotropic associations. Existing selection methods examine pleiotropic associations one by one at a scale of either the genetic variant or the gene, and thus cannot consider all the genetic information at the same time. To address this limitation, we propose a new approach called MPSG (Meta-analysis model adapted for Pleiotropy Selection with Group structure). This method performs a penalized multivariate meta-analysis method adapted for pleiotropy and takes into account the group structure information nested in the data to select relevant variants and genes (or pathways) from all the genetic information. To do so, we implemented an alternating direction method of multipliers algorithm. We compared the performance of the method with other benchmark meta-analysis approaches such as GCPBayes, PLACO, and ASSET by considering as inputs different kinds of summary statistics. We provide an application of our method to the identification of potential pleiotropic genes between breast and thyroid cancers.

PMID:41241933 | DOI:10.1093/biostatistics/kxaf037