Category: Statistics

Health Justice. 2025 Aug 16;13(1):52. doi: 10.1186/s40352-025-00362-4.

ABSTRACT

BACKGROUND: We describe a longitudinal mixed methods program evaluation protocol for a novel peer-led housing program for older men transitioning from prison to the community after completing long sentences of incarceration. The program departs from traditional community corrections models by providing peer-run housing designed to build and enhance peer and community social ties. This previously untested program relies on the principles of network alteration and provides a case study for examining interpersonal mechanisms underlying behavioral health and justice related outcomes.

METHODS: We use mixed methods and longitudinal social network analysis to evaluate the program, while also applying implementation science to document program development. We focus our evaluation on key health and social outcomes, including mental health, substance use, stress, health risk behaviors, well-being, financial security, housing, and recidivism. With longitudinal surveys, we collect (1) dynamic network data of resident and staff relationships and (2) behavioral health/social data of participants. We also administer longitudinal resident and staff interviews. Resident interviews focus on interpersonal relationships and reentry experiences, while staff interviews describe program implementation. We apply longitudinal statistical models to complete (i.e., sociocentric) network data within the house to examine how dynamic network properties connect to changes in residents’ health, behavioral, and social outcomes. We integrate longitudinal survey, individual-level (i.e., egocentric) network, and qualitative data to understand how the program works. To evaluate program impacts for long-term health and social outcomes, we use an untreated matched sample to compare 6- and 12-months post-prison release outcomes using administrative data related to rearrest/reincarceration and behavioral health.

PRE-IMPLEMENTATION RESULTS: We use a logic model to present and organize pre-implementation results from interviews with program staff and peer mentors. Our results describe program design and intended goals, while highlighting how the program is rooted in principles of peer support, trauma-informed care, and restorative justice to address unique stressors of incarceration to foster responsibility and facilitate reintegration.

DISCUSSION: Community program evaluation research allows us to document real-world contextual factors that may drive intervention effectiveness. Results of the mixed methods evaluation will provide a comprehensive understanding of one network-based program’s ability to support health and social outcomes of older, previously incarcerated men. Results may inform future reentry services.

PMID:40817958 | DOI:10.1186/s40352-025-00362-4

Eur Radiol. 2025 Aug 16. doi: 10.1007/s00330-025-11898-2. Online ahead of print.

ABSTRACT

OBJECTIVES: We aimed to evaluate the diagnostic performance of deep learning (DL)-based radiomics models for the noninvasive prediction of isocitrate dehydrogenase (IDH) mutation and 1p/19q co-deletion status in glioma patients using MRI sequences, and to identify methodological factors influencing accuracy and generalizability.

MATERIALS AND METHODS: Following PRISMA guidelines, we systematically searched major databases (PubMed, Scopus, Embase, Web of Science, and Google Scholar) up to March 2025, screening studies that utilized DL to predict IDH and 1p/19q co-deletion status from MRI data. We assessed study quality and risk of bias using the Radiomics Quality Score and the QUADAS-2 tool. Our meta-analysis employed a bivariate model to compute pooled sensitivity and specificity, and meta-regression to assess interstudy heterogeneity.

RESULTS: Among the 1517 unique publications, 104 were included in the qualitative synthesis, and 72 underwent meta-analysis. Pooled estimates for IDH prediction in test cohorts yielded a sensitivity of 0.80 (95% CI: 0.77-0.83) and specificity of 0.85 (95% CI: 0.81-0.87). For 1p/19q co-deletion, sensitivity was 0.75 (95% CI: 0.65-0.82) and specificity was 0.82 (95% CI: 0.75-0.88). Meta-regression identified the tumor segmentation method and the extent of DL integration into the radiomics pipeline as significant contributors to interstudy variability.

CONCLUSION: Although DL models demonstrate strong potential for noninvasive molecular classification of gliomas, clinical translation requires several critical steps: harmonization of multi-center MRI data using techniques such as histogram matching and DL-based style transfer; adoption of standardized and automated segmentation protocols; extensive multi-center external validation; and prospective clinical validation.

KEY POINTS: Question Can DL based radiomics using routine MRI noninvasively predict IDH mutation and 1p/19q co-deletion status in gliomas, and what factors affect diagnostic accuracy? Findings Meta-analysis showed 80% sensitivity and 85% specificity for predicting IDH mutation, and 75% sensitivity and 82% specificity for 1p/19q co-deletion status. Clinical relevance MRI-based DL models demonstrate clinically useful accuracy for noninvasive glioma molecular classification, but data harmonization, standardized automated segmentation, and rigorous multi-center external validation are essential for clinical adoption.

PMID:40817944 | DOI:10.1007/s00330-025-11898-2

Biomed Microdevices. 2025 Aug 16;27(3):39. doi: 10.1007/s10544-025-00765-y.

ABSTRACT

The urgent need for adenosine triphosphate (ATP) detection spans various fields, particularly in biology and medicine. Developing a simple, quick, label-free, and highly sensitive biosensor for ATP detection is crucial. In this study, we created a label-free biosensor using a field-effect device, specifically an electrolyte-insulator-semiconductor (EIS) sensor, which was functionalized with aptamer and graphene. We prepared a nanocomplex by combining graphene with bovine serum albumin (BSA) in PBS and subjecting it to ultrasonication. This Graphene/BSA mixture was then combined with 70% glutaraldehyde to form the Graphene/BSA/GA nanocomplex. The successful modification of the EIS biosensor surface with Graphene/BSA/GA and aptamer immobilization was confirmed using atomic force microscopy (AFM), which indicated successful molecule attachment through surface roughness. Electrochemical characterization revealed that the biosensor is sensitive to ATP concentrations ranging from 0.1 nM to 100 nM, with a detection limit as low as 0.32 nM. Statistical analysis demonstrated the biosensor’s high sensitivity and specificity for ATP. Furthermore, the biosensor maintained stable performance for ATP detection over a period of 5 days. This sensing approach effectively detected ATP with outstanding performance, showing significant potential for advancing label-free ATP detection technologies.

PMID:40817942 | DOI:10.1007/s10544-025-00765-y

J Neural Transm (Vienna). 2025 Aug 16. doi: 10.1007/s00702-025-02999-9. Online ahead of print.

ABSTRACT

Precision medicine driven by clinical biomarkers is the state-of-art management approach for Parkinson’s disease (PD). Whether pattern of striatal dopaminergic deficiency (demonstrated by single-photon emission CT (SPECT) scanning with ¹²³I-Ioflupane, DaTSCAN) could be a biomarker predicting levodopa requirement in early PD is not known. Participants with early PD (disease duration (DD) ≤ 5 years, Hoehn and Yahr (H&Y) ≤ 3) who underwent DaTSCAN as a part of clinical-diagnostic work up and were enrolled in the “Non-motor Longitudinal International Study” (UK National Institute for Health Research Clinical Research Network Number 10084) were included in this cross-sectional analysis. Specific DaTSCAN binding ratios were analyzed for each striatum, caudate nucleus and putamen and the z-scores were derived normalizing the images to age and gender-matched healthy controls from the European-Database-of-DaTSCAN-of-healthy-controls (ENC-DAT). Using linear regression analysis, we explored the impact of DaT-uptake z-scores for more severely affected putamen, caudate nucleus and striatum on the LEDD. Statistically significant predictors identified in the univariable analysis were included in the multivariable analysis with DD and H&Y as additional independent variables. 43 PwP (30% female; age: 61.91 ± 11.45years; DD: 2(0-5) years; H&Y: 2(1-3); LEDD: 424.27 ± 342.62 mg) were assessed 19.12 ± 13.11 months following the DaTSCAN. In a multivariable linear regression analysis, when adjusted for DD and H&Y, z-caudate nucleus (B=-134.073, 95% CI -262.715 – -5.431, p = 0.042) and z-striatum (B=-162.137, 95% CI -306.306 – -17.967, p = 0.028), were statistically significant predictors of LEDD, while z-putamen was not (p = 0.086). In early PD, striatal DaT-uptake z-scores may serve as biomarkers that could aid the LEDD estimation and guide treatment decisions towards personalized care.

PMID:40817940 | DOI:10.1007/s00702-025-02999-9

Neurosurg Rev. 2025 Aug 16;48(1):605. doi: 10.1007/s10143-025-03760-2.

ABSTRACT

Double anastomosis Superficial Temporal Artery-Middle Cerebral Artery (STA-MCA) bypass in Moyamoya disease (MMD) and cerebral atherosclerosis disease (CAD) remains underexplored. We aim to evaluate the clinical efficacy and vascular dynamics of double anastomosis STA-MCA bypass in each group, MMD and CAD. A systematic search was conducted in ScienceDirect, Embase, PubMed, and Cochrane databases until September 2024. Efficacy outcomes included stroke occurrences, patency, modified Rankin Scale (mRS), and complications. Vascular parameters included cerebral blood flow (CBF), cut flow index (CFI), MCA perfusion, and mean transit time (MTT). The meta-analysis included 15 studies with 417 patients (46.18 ± 4.69 years). At final follow-up, the MMD group showed lower total stroke (12 ± 7% vs. 20 ± 9%; p = 0.63), ischemic stroke (8 ± 5% vs. 18 ± 8%; p = 0.26), hemorrhagic stroke (4 ± 3% vs. 10 ± 7%; p = 0.49), and patency (85 ± 11% vs. 98 ± 8%; p = 0.10) compared to CAD group, without statistical difference. The MMD group exhibited reduced mRS (MD: -0.57; 95%CI: -0.95 to -0.19; p = 0.003), increased CBF (MD: 14.75; 95%CI: 2.32 to 27.19; p = 0.020), and increased MCA perfusion (MD: 9.28; 95%CI: 1.19 to 17.37; p = 0.025). CFI (1.30 ± 0.11 vs. 1.08 ± 0.13; p = 0.02) and MTT (4.49 ± 0.5 vs. 1.27 ± 0.27; p < 0.001) were significantly higher in the CAD group. Both 1D2R (MD: 27.59; 95%CI: 4.88 to 50.30; p = 0.017) and 2D2R (MD: 14.75; 95%CI: 2.32 to 27.19; p = 0.020) techniques exhibited increase in CBF. Additionally, double anastomosis showed a higher CFI compared to single anastomosis (MD: 0.25; 95%CI: 0.12-0.38; p < 0.001). Double anastomosis STA-MCA bypass demonstrated comparable stroke and patency rates between groups. MMD patients had significantly higher CFI and MTT values, with double anastomosis yielding superior CFI compared to single anastomosis.

PMID:40817938 | DOI:10.1007/s10143-025-03760-2

Diabetologia. 2025 Aug 16. doi: 10.1007/s00125-025-06517-0. Online ahead of print.

ABSTRACT

AIMS/HYPOTHESIS: There is no clear consensus regarding accurate risk stratification in early pregnancy for later developing gestational diabetes mellitus (GDM). Therefore, this study aims to evaluate the predictive performance of an OGTT and several biomarkers in the first trimester of pregnancy. Their association with insulin action, beta cell function and requirement for insulin were additionally assessed.

METHODS: In this prospective cohort study, we included 657 pregnant women in six Central European centres. Patient history and anthropometric data were obtained, a blinded 75 g OGTT was performed and biochemical markers were assessed at a median gestational age of 13.4 weeks (IQR 12.7-14.1). Another OGTT was performed in later pregnancy to identify women with GDM. A detailed investigation of glucose homeostasis was performed at both visits in a subgroup of women.

RESULTS: Eighty-three women (12.6%) developed GDM. Progression to GDM was fairly well predicted by glucose concentrations during the early OGTT in terms of areas under the receiver operating characteristic curves (OGTT glucose at fasting: 0.68; OGTT glucose at 60 min: 0.74; OGTT glucose at 120 min: 0.72). Some biomarkers showed significant but modest predictive accuracy. Early gestational OGTT glucose concentrations were further associated with impaired insulin sensitivity and beta cell dysfunction, as well as the requirement for insulin in later pregnancy.

CONCLUSIONS/INTERPRETATION: Although the accurate diagnosis of GDM before 24 weeks remains an ongoing discussion, dynamically assessed glucose concentrations during an early OGTT were closely associated with impaired glucose homeostasis and showed good predictive accuracy for later development of GDM as well as the requirement for insulin. These findings may be used to develop a protocol to distinguish between low- and high-risk mothers. Trial registration ClinicalTrials.gov NCT02035059.

PMID:40817933 | DOI:10.1007/s00125-025-06517-0

Clin Chem Lab Med. 2025 Aug 18. doi: 10.1515/cclm-2025-0763. Online ahead of print.

ABSTRACT

Laboratory data can be meaningful only when compared with reliable reference data; therefore, the estimation of reliable reference data is just as important as the accurate measurement of measurands in patient samples. Since analyte concentrations in the human body are influenced by both random variations (such as biological fluctuations) and systematic variations (such as physiological rhythms and age-related changes), the conventional model for estimating reference data – based solely on the statistical distribution of single-sample measurements from reference individuals – may not provide sufficiently reliable information for interpreting patient results. Therefore, a paradigm shift from relying solely on single-sample measurement distributions to incorporating metabolic changes observed in the human body when estimating reference intervals may enhance the clinical value of laboratory data for an effective clinical decision making and patient care. This opinion paper aims to summarize how to facilitate this transition and to identify the most suitable model for estimating reference intervals that reflect underlying metabolic dynamics.

PMID:40817855 | DOI:10.1515/cclm-2025-0763

Anim Genet. 2025 Aug;56(4):e70037. doi: 10.1111/age.70037.

ABSTRACT

With advancing genomic technologies, single-nucleotide polymorphism (SNP) arrays and whole genome sequencing (WGS) have become essential tools in equine genetic research. In this study, we assessed the concordance in SNP calls and trait-mapping efficacy by comparing data of 21 horses both genotyped on the Equine 670 K SNP array and sequenced at either ~12× or ~30× depth. Our analysis revealed that higher sequencing depths were significantly associated with fewer discordant calls between platforms. Additionally, we investigated the most frequent no-call and discordant positions and identified positions that were indels or multiallelic in the WGS. To assess the effectiveness of the 670 K SNP array vs. WGS in trait association studies, we mapped the chestnut coat color. Both technologies showed a clear peak at the expected locus, although neither association had loci reaching Bonferroni-corrected statistical significance, which was not statistically possible in this small group of horses. The findings of this study provide valuable insights for making informed decisions when selecting between SNP arrays and WGS at varying sequencing depths for equine genomic research applications.

PMID:40817846 | DOI:10.1111/age.70037

J Clin Endocrinol Metab. 2025 Aug 16:dgaf461. doi: 10.1210/clinem/dgaf461. Online ahead of print.

ABSTRACT

OBJECTIVE: To study bone turnover markers (BTM) and sex steroids in relation to Trabecular Bone Score (TBS) in men.

MATERIAL AND METHODS: Longitudinal, population-based study in 465 healthy men, aged 25-45 years at baseline. Lumbar spine TBS was calculated with TBS iNsight® version 4 (v19.4.1, core module, Medimaps, Pessac, France), which adjusts for soft tissue thickness correction via DXA measurements. Sex hormone binding globulin (SHBG), C-terminal telopeptide and pro collagen type 1 N-terminal propeptide were measured using immunoassays. Total testosterone (T) and estradiol (E2) were determined by liquid chromatography-tandem mass spectrometry, free T and free E2 calculated. Statistical analyses were conducted via linear mixed-effects modelling.

RESULTS: At baseline, TBS was positively associated with free testosterone (p=0.01), free estradiol and total estradiol (both p<0.001), but not with total testosterone nor with BTM. Over a follow-up of 12.5 years, TBS declined by 1.43% (p<0.001). Higher baseline BMI and trunk fat were predictive of greater decreases in TBS (p=0.01 and p=0.02). Baseline levels of sex steroids and BTM nor changes therein were associated with changes in TBS.

CONCLUSIONS: TBS already decreases in young and middle-aged healthy men, corroborating earlier studies showing early decrease of trabecular bone volume and changes in trabecular microarchitecture. Although we identified some potentially contributing determinants, the underlying mechanisms of changes in TBS and trabecular bone in young men are yet to be fully elucidated.

PMID:40817832 | DOI:10.1210/clinem/dgaf461

J Vet Diagn Invest. 2025 Aug 16:10406387251357233. doi: 10.1177/10406387251357233. Online ahead of print.

ABSTRACT

Congestive heart failure (CHF) in feedyard cattle is of increasing concern among producers and can be difficult to diagnose definitively postmortem. In a cross-sectional observational study, we evaluated gross pathology findings, various heart measurements, and subjective heart scores (1-5 scale: 1 = normal, 5 = severely remodeled) to identify heart disease postmortem. In postmortem examinations of 346 feedyard deaths, we classified 106 (30.6%) cases as cardiac enlargement or misshapen ventricle (CEMV) when there was an abnormal heart shape or dilated ventricle(s), and no signs of infectious heart disease. CHF was defined as a CEMV case with chronic passive congestion of the liver (i.e., nutmeg liver) and ≥2 of the following lesions: serous or serosanguineous pleural, pericardial, or peritoneal effusion. Eleven of the 346 autopsied cattle were classified as having CHF. Descriptive statistics and multivariate models were used to identify statistical associations between objective heart measurements or subjective heart scores and the prevalence of CEMV or CHF. CEMV cases had significantly increased heart widths, thinner left ventricular free walls, and expanded right ventricular lumen areas (p <0.05). The CHF model did not converge because we had too few cases to be able to evaluate associations between CHF and variables of interest.

PMID:40817806 | DOI:10.1177/10406387251357233