Category: Statistics

Eur Radiol. 2026 Jan 24. doi: 10.1007/s00330-025-12303-8. Online ahead of print.

ABSTRACT

OBJECTIVE: To evaluate the performance of model-based dynamic contrast-enhanced (DCE)-MRI and diffusion-weighted imaging (DWI) in determining the disease activity of thyroid-associated ophthalmopathy (TAO), and to establish their additional value to fat-suppressed T2-weighted imaging (FS-T2WI) for staging TAO.

MATERIALS AND METHODS: Seventy-two patients with TAO (48 active, 96 eyes; 24 inactive, 48 eyes) were prospectively enrolled. DCE-MRI, DWI and FS-T2WI were scanned for pre-treatment evaluation. Simplified histogram parameters (min, mean, max) of DCE-MRI-derived K^trans, K_ep and V_e, apparent diffusion coefficient (ADC) and signal intensity ratio (SIR) on FS-T2WI of extraocular muscles were calculated for each orbit and compared between active and inactive groups. Multivariate analyses were used to identify independent indicators for disease activity. Receiver operating characteristic (ROC) curves analyses and DeLong tests were performed to evaluate and compare the performances of the identified significant imaging parameters and their combinations.

RESULTS: Active TAO patients showed significantly higher mean and maximum V_e, higher minimum, mean and maximum ADC, higher minimum, mean and maximum SIR than inactive patients (p < 0.05). Mean SIR (odds ratio (OR) = 3.449, p = 0.002), mean ADC (OR = 1.008, p < 0.001), and mean V_e (OR = 14.138, p = 0.022) were found to be independent predictors of active TAO. Combination of mean V_e, mean ADC and mean SIR outperformed mean SIR alone in staging TAO (area under ROC curves, 0.839 vs 0.769, p = 0.016).

CONCLUSION: DCE-MRI and DWI could determine the disease activity of TAO and provide additional value to FS-T2WI in staging TAO.

KEY POINTS: Question Fat-suppressed T2-weighted imaging was the most commonly used imaging technique for determining the disease activity of thyroid-associated ophthalmopathy; however, its performance needs to be improved. Findings Dynamic contrast-enhanced magnetic resonance imaging and diffusion-weighted imaging could provide added value to fat-suppressed T2-weighted imaging for determining the clinical activity of thyroid-associated ophthalmopathy. Clinical relevance Dynamic contrast-enhanced magnetic resonance imaging and diffusion-weighted imaging can provide information about tissue permeability and water molecule diffusion of extraocular muscles in patients with thyroid-associated ophthalmopathy (TAO), and therefore provide additional value to fat-suppressed T2-weighted imaging in staging TAO.

PMID:41578080 | DOI:10.1007/s00330-025-12303-8

Behav Res Methods. 2026 Jan 23;58(2):50. doi: 10.3758/s13428-025-02940-8.

ABSTRACT

Categorization studies, in which stimuli vary along a category continuum, are becoming increasingly popular in psychological science. These studies demonstrate the effect of category ambiguity on various behavioral and neural measures. In such studies, researchers manipulate objective category levels by varying the physical properties of the stimuli, and then use these levels as predictors of behavior-assuming they map directly onto participants’ perceived locations along the category continuum. This approach might not be optimal, considering the variability in participants’ category boundary locations (their point of subjective equality, or PSE). In this tutorial, we propose addressing this issue by estimating participants’ individual points of subjective equality, adjusting category levels relative to these points, and conducting statistical analyses on the subjective category levels. Implementing this method significantly improves the statistical power of the analysis in both experimental and simulated data. Adjusting stimulus levels by the points of subjective equality is highly suited for social categorization studies, in which points of subjective equality vary significantly. On a broader scale, it can be applied to a variety of categorization, discrimination, and decision-making studies.

PMID:41578073 | DOI:10.3758/s13428-025-02940-8

Neurosurg Rev. 2026 Jan 24;49(1):154. doi: 10.1007/s10143-025-04075-y.

ABSTRACT

Trigeminal neuralgia (TN) is a neuropathic pain disorder characterized by severe orofacial pain. The underlying pathophysiological nuances remain under study, and their understanding is key to developing new and more effective therapies for this debilitating disease. The field of cellular therapies for neurological diseases is continuously evolving. This systematic review was performed after the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. We sought studies on the preclinical and clinical uses of cellular therapies for trigeminal neuralgia. We included 8 studies encompassing 1 clinical and 7 preclinical applications. Of the preclinical studies, four used Stem Cells from Human Exfoliated Deciduous Teeth (SHED), two used Olfactory Ensheathing Cells (OECs), and one used Bone Marrow Mesenchymal Stem Cells (BMSCs). All preclinical studies showed a statistically significant behavioral improvement in groups receiving cellular therapies compared to controls (p < 0.05), with either local or systemic delivery. In addition, cellular therapies have the potential to mitigate TN by promoting myelin repair, reducing neuroinflammation, and modulating pain-related pathways. The only clinical report in the literature described the incidental use of Adipose-Derived Stem Cells (ADSCs) during a facial cosmetic procedure in a 60-year-old female with a long history of TN, who remained pain-free at 2-year follow-up. These findings highlight the promising role of cellular therapies in the treatment of TN, demonstrating significant behavioral and molecular benefits in preclinical models and a compelling clinical case. Further rigorous clinical studies are necessary to establish their safety, efficacy, and long-term therapeutic impact.

PMID:41578017 | DOI:10.1007/s10143-025-04075-y

Unfallchirurgie (Heidelb). 2026 Jan 23. doi: 10.1007/s00113-026-01678-y. Online ahead of print.

ABSTRACT

BACKGROUND: The treatment of fragility fractures of the pelvis (FFP) is becoming increasingly more important due to the demographic changes. Older age and the associated multimorbidity pose a challenge for the optimal treatment of pelvic fractures.

METHOD: A total of 36 patients were included. They were categorized according to the FFP classification with the following distribution: 44.4% FFP II, 16.7% FFP III and 38.9% FFP IV. Transiliosacral screws used unilaterally/bilaterally (cannulated 7.3 mm titanium screws with 32 mm thread, MedTech J&J, Umkirch, Germany) were compared to a continuous transiliosacral sacral rod (cannulated 7.5 mm rod, Marquardt, Spachingen, Germany) for stabilization of the posterior pelvic ring (screw vs. rod) and in combination with or without a ventral supra-acetabular external fixator (steel Schanz screws with cross-connectors classified as ⌀Fix vs. Fix). The parameters measured were the duration of surgery, complication and revision rates and functional outcomes. The quality of life on the visual analogue scale (EQ-VAS), EQ-5D-5L index and Elderly Mobility Scale (EMS) were compared 1 day, 6 weeks and 6 months postoperatively.

RESULTS: The mean age of the overall cohort was 81.5 ± 7.7 years and the overall mortality rate was 5.6%. The higher complication (29.4% vs. 17.6%, p = 0.419) and revision rates (5.9% vs. 0.0%, p = 0.310) of the screw group were not statistically significant. The EQ-VAS, EQ-5D-5L and EMS showed no significant differences between screw vs. rod. The fix group had a longer operation time (47.2 ± 9.2 min vs. 35.2 ± 20.2 min, p = 0.005) but a lower complication rate (11.1% vs. 28.0%, p = 0.306); however, their mobility was significantly reduced postoperatively and after 6 weeks (EMS day 1: 6 ± 4 vs. 11 ± 4, p = 0.003; week 6: 12 ± 2 vs. 16 ± 3, p = 0.010).

DISCUSSION: Osteoporosis treatment had not been performed prior to the injury in 64.7% of the injured patients, 29.4% received a basic treatment for osteoporosis and 5.9% received specific treatment for osteoporosis. Surgical treatment of FFP II-IV showed a good clinical outcome with age-appropriate values after 6 months. The clinical outcome was the same after 6 months regardless of the surgical procedure.

PMID:41578015 | DOI:10.1007/s00113-026-01678-y

Environ Monit Assess. 2026 Jan 23;198(2):169. doi: 10.1007/s10661-026-14993-z.

ABSTRACT

As a crucial freshwater resource, groundwater plays an indispensable role in arid and semi-arid regions characterized by low annual precipitation and frequent droughts. Developing computational frameworks for groundwater level prediction is essential to advance sustainable water resource management. This study proposes a hybrid deep learning model (CNN-GRU-Attention) for groundwater depth forecasting, integrating convolutional neural networks (CNN), gated recurrent units (GRU), and attention mechanisms. The methodological framework commenced with a spatiotemporal analysis of groundwater depth dynamics in Zhengzhou, China. Subsequently, multiple machine learning and deep learning algorithms were systematically evaluated to predict groundwater depth using four input variables: monthly evaporation, precipitation, average temperature, and groundwater extraction. These variables were rigorously selected through the Shannon entropy method. Model performance was quantified using three statistical metrics: MAE, RMSE, and R². Results indicate that the CNN-GRU-Attention model demonstrates superior performance in groundwater depth forecasting, achieving MAE values of 0.4-0.5, RMSE values of 0.5-0.6, and R² values of 0.8-0.9. To fully evaluate the performance of the model, we designed two hypothetical scenarios. First, we analyzed changes in the model’s predictive performance under conditions of reduced data, when the data volume is reduced by 10-25%, the CNN-GRU-Attention model still outperforms other models in predictive performance. Second, to maintain stable groundwater depth under drought-induced rainfall reduction conditions, controlled extraction measures should be implemented to balance recharge and withdrawal. Under this special rainfall scenario, a reduction in extraction volume of 42 million m³ is more conducive to maintaining groundwater stability. This model provides an effective predictive framework and offers valuable insights for sustainable groundwater management in arid regions.

PMID:41577970 | DOI:10.1007/s10661-026-14993-z

Soc Sci Med. 2026 Jan 16;393:118995. doi: 10.1016/j.socscimed.2026.118995. Online ahead of print.

ABSTRACT

Ethnoracial segregation has been associated with worse cognitive functioning among Black older adults, while its impact on Latinx individuals is less clear. We investigated whether Black and Latinx older adults living in segregated neighborhoods demonstrate worse magnetic resonance imaging (MRI) outcomes. We used data on participants from the University of California Davis Alzheimer’s Disease Research Center. MRI outcomes included hippocampal and white matter hyperintensity (WMH) volumes. Black and Latinx segregation was defined using the Getis-Ord (Gi∗) statistic, which compares the proportion of Black or Latinx residents, respectively, in the participant’s Census tract to surrounding neighborhoods and greater study region (higher Gi∗ = greater clustering/segregation). Multivariable linear regression analyses examined associations between Gi∗ segregation measures and MRI outcomes, stratified by the participants’ ethnoracial group (Black, Latinx, or White). Participants (n = 269) were on average 74 ± 7 years of age and 24 % were Black, 25 % were Latinx, and 51 % were White. In adjusted analyses, Black participants in more Latinx segregated neighborhoods had lower hippocampal volumes, and Latinx participants in more Black segregated neighborhoods had lower hippocampal volumes. Latinx participants in more Latinx segregated neighborhoods had greater white matter hyperintensity volumes. Overall, Black and Latinx but not White participants living in segregated neighborhoods had worse MRI outcomes. Future studies are needed to replicate our findings in geographically diverse samples and to elucidate the potential psychosocial/social determinant and biological mechanisms that relate segregation to brain health (e.g., Latinx segregated neighborhoods may have fewer recreational and physical activity resources to promote healthy lifestyles).

PMID:41576475 | DOI:10.1016/j.socscimed.2026.118995

Reprod Biol. 2026 Jan 21;26(2):101179. doi: 10.1016/j.repbio.2026.101179. Online ahead of print.

ABSTRACT

Fetal growth restriction (FGR) is a severe pregnancy complication often caused by placental insufficiency. Proper trophoblast invasion is essential for placental development and function, ensuring adequate nutrient and oxygen supply to the developing fetus. Dysregulation impairs placental perfusion, leading to FGR. This study uses a calorie-restricted mouse model to investigate genes/molecular mechanisms regulating trophoblast invasion across gestational timepoints. Pregnant mice received either a standard or 50 % calorie-restricted diet from E8.5. Placentas and invasion sites were analyzed at E10.5, E12.5, E14.5, E16.5, and E17.5. mRNA sequencing and RT/qPCR examined trophoblast invasion-related genes (Mmp2, Mmp9, Efna1, Rac1, Rras, Ascl2, Tfap2c, Prl7b1) and angiogenesis genes (Vegfa, Vegfb, Pdgf, Akt3). Immunohistochemistry of trophoblast cells (cytokeratin 8, CK8) and endothelial cell markers (endomucin, CD31, CCD105, VEGFR2) was performed. Statistical analysis used Student’s t-test. Caloric restriction significantly reduced fetal/placental weights from E12.5, with persistent growth restriction at E16.5, and E17.5. IHC at E17.5 showed reduced decidual depth, trophoblast invasion distance, and trophoblast quantity within the decidua. This impaired growth was accompanied by reduced expression of trophoblast invasion genes (Mmp2, Mmp9, Efna1, Rac1, Rras, Ascl2, Tfap2c, Prl7b1) in FGR placentas, with a reduction in CK8 trophoblast staining. Angiogenesis reduction in FGR was demonstrated with reduced Vegfa, Vegfb, and Akt3 and supported by reduced CD31, CD105, and VEGF2 endothelial cell markers A caloric-restriction mouse model replicates key FGR pathophysiology, including reduced fetal/placental growth, downregulation of trophoblast invasion genes, impaired trophoblast invasion into the decidua, and reduced placenta angiogenesis. These findings offer molecular insights into placental insufficiency that merits further exploration regarding FGR pathogenesis.

PMID:41576455 | DOI:10.1016/j.repbio.2026.101179

J Hazard Mater. 2026 Jan 17;503:141192. doi: 10.1016/j.jhazmat.2026.141192. Online ahead of print.

ABSTRACT

The design and development of a universal detection system for toxic chemicals such as chemical warfare (CW) agents and pesticides offers a promising solution for safety and surveillance in defense, environment, and health sectors. The ability to detect a wide spectrum of these hazardous chemicals rapidly, simply, and cost-effectively through a visible color change provides a highly practical and impactful tool. This manuscript introduces a universal platform that enables the detection of nerve agents, blister agents, and organophosphorus (OP) pesticides using 5,5′-dithiobis-(2-nitrobenzoic acid) (DTNB) as a central probe through two distinct strategies. For nerve agent detection, 1-phenylbutane-1,2,3-trione-2-oxime (1) reacts with nerve agents to form an intermediate, phosphorylated oxime (5). This intermediate rapidly decomposes, releasing cyanide ions that subsequently react with DTNB to produce a ‘turn-on’ response. Blister agents are identified through their rapid reaction with sodium thiosulfate at room temperature, forming Bunte salts that do not interact with DTNB, resulting in a ‘turn-off’ response. Beyond nerve and blister agents, the applicability of this strategy was further expanded to detect OP pesticides, highlighting its broad-spectrum potential. The approach effectively overcomes the challenge of achieving high specificity amid potentially cross-reactive substances. Moreover, this platform also demonstrated a robust performance across diverse matrices, including soil, water, and fruit. Recovery experiments in soil showed acceptable precision, underscoring both the reliability of the method and its environmental relevance. To facilitate real-time field deployment for first responders, a portable sensor kit was fabricated to visually detect CW agents and OP pesticides. Smartphone-assisted colorimetric analysis of a detector paper delivered reliable analytical performance, exhibiting statistically validated sensitivity and reproducible responses.

PMID:41576450 | DOI:10.1016/j.jhazmat.2026.141192

J Public Health Manag Pract. 2026 Mar-Apr 01;32(2):260-267. doi: 10.1097/PHH.0000000000002284. Epub 2025 Nov 24.

ABSTRACT

CONTEXT: Public health organizations are increasingly recognizing the value and potential of data science. However, a gap remains in understanding how data science is being applied in public health.

OBJECTIVE: This article provides a comprehensive overview of data science applications in real-world public health settings. By describing the characteristics of projects supported by the Centers for Disease Control and Prevention’s Data Science Upskilling (DSU) program during 2019-2023, we seek to guide future efforts in public health data science workforce development and data modernization.

METHODS: We manually reviewed DSU applications and final presentations about the projects compiled during 2019-2023. We analyzed projects based on 7 characteristics, including public health domain and task, data science topic and method, data modality, tools, and programming languages used.

RESULTS: DSU supported 112 data science projects across 5 annual cohorts (2019-2023). Many projects addressed the COVID-19 pandemic (13%), infectious diseases (13%), and vaccines (11%). Approximately half the projects used data visualization (54%) and statistics (51%), with 42% employing artificial intelligence (AI) and machine learning (ML). Furthermore, 52% of projects were designed to support decision making, and 22% sought to improve processes and programs. Learners primarily used RStudio (50%), Jupyter Notebooks (41%), and Power BI (26%), along with Python (56%) and R (55%). AI and ML use increased from 33% of projects in 2019 to 56% in 2023, demonstrating an evolving focus on advanced methodologies.

CONCLUSIONS: Many teams prioritized data visualization, such as dashboards and visualization tools to support decision making, indicating opportunities for additional infrastructure and training in this area. We observed increasing use of AI and ML, suggesting a need for staff upskilling in these domains. Optimally leveraging data science technologies will require workforce development strategies and data modernization efforts to keep pace with the rapidly evolving field.

PMID:41576408 | DOI:10.1097/PHH.0000000000002284

J Public Health Manag Pract. 2026 Mar-Apr 01;32(2):179-190. doi: 10.1097/PHH.0000000000002290. Epub 2025 Nov 24.

ABSTRACT

CONTEXT: Vision Zero (VZ) is a road safety initiative that seeks to address the problem of road fatalities using a Safe System approach, a holistic endeavor embedded in public health principles that seeks to build layers of protection across transportation systems to eliminate road fatalities and serious injuries. Since 2020, a multidisciplinary research team established a statewide collaborative to support communities pursuing VZ initiatives across North Carolina.

PROGRAM: The North Carolina VZ collaborative “support model” was created to meet the need for community-based, multisector efforts using a Safe System approach. The support model aims to increase community capacity to more effectively build cross-disciplinary coalitions, pool needed resources, and strengthen adaptive leadership skills to reduce roadway fatalities.

IMPLEMENTATION: The support model approach is used to engage communities in building skills in cross-sector collaboration, adaptive leadership, and evidence-based safety procedures. This is accomplished through structured monthly touchpoint meetings with small groups of community partners for peer learning, quarterly “all-hands” meetings to coordinate efforts across the state and provide resources, and an annual team-based multiday Leadership Institute.

EVALUATION: From 2020 to 2025, there was notable growth in community participation, from 7 to 33 communities. Of communities with more than 1 year of participation (n = 19), more than half advanced VZ implementation with communities moving from an exploration stage to an installation (n = 8) or initial implementation (n = 2) stage. In 2023, interviews with partner community leads (n = 15) indicated that partners utilized resources provided, applied skills they learned at the Leadership Institute, benefited from the peer network, and identified opportunities for increasing the benefits of the support model.

DISCUSSION: The support model demonstrates a promising practice for increasing capacity building and cross-sector collaboration for road safety initiatives requiring complex systems change such as VZ.

PMID:41576406 | DOI:10.1097/PHH.0000000000002290