Tag: statistics

BMC Musculoskelet Disord. 2026 Jan 2. doi: 10.1186/s12891-025-09462-6. Online ahead of print.

ABSTRACT

PURPOSE: Thoracic hyperkyphosis (THK) is a prevalent spinal deformity in adolescents, which can impair respiratory function by affecting diaphragm morphology and thoracic mechanics. This study investigates the relationship between THK and diaphragmatic thickness using ultrasonography (USG) to explore its diagnostic and therapeutic implications.

METHODS: This prospective study included 25 adolescents with THK and 29 healthy controls. Diaphragmatic thickness was measured using ultrasonography at the end of inspiration and expiration, and spirometry was conducted to assess respiratory function. Thoracic hyperkyphosis was diagnosed based on Cobb angle measurements. Statistical analysis included regression and ROC analyses to determine predictors of THK and evaluate diagnostic accuracy. The study was approved by the local Ethics Committee of the University of Healthy Science, Gaziosmanpasa Training and Research Hospital (IRB number: 109, date: 16 August 2023). The study protocol was registered to the clinicaltrial.gov.tr (NCT06418334 and initial release date: 01/11/2024).

RESULTS: Adolescents with THK exhibited significantly higher Cobb angles (59.3 ± 7.9°) compared to controls (34.6 ± 3.7°, p < 0.001). Diaphragmatic thickness during inspiration (2.5 mm vs. 3.1 mm, p = 0.125) and expiration (1.50 mm vs. 1.56 mm, p = 0.965) was lower in the THK group, with significant reductions in thickening ratio (1.76 ± 0.29 vs. 1.98 ± 0.30, p = 0.008) and thickening fraction (DTf) (75.4% vs. 97.6%, p = 0.008). A negative correlation was observed between Cobb angle and diaphragmatic thickening (r=-0.378, p = 0.005). ROC analysis indicated that USG parameters, particularly thickening fraction (AUC = 0.72, p = 0.002), are reliable predictors of THK.

CONCLUSION: Adolescents with THK demonstrate reduced diaphragmatic thickness and impaired respiratory function. Ultrasonography emerges as a valuable, non-invasive tool for early diagnosis and monitoring of THK, offering potential for timely intervention to mitigate respiratory complications.

PMID:41484972 | DOI:10.1186/s12891-025-09462-6

J Health Popul Nutr. 2026 Jan 3. doi: 10.1186/s41043-025-01211-8. Online ahead of print.

ABSTRACT

BACKGROUND: Preterm birth (PTB) is a global epidemic, defined as delivery before 37 weeks of gestation, and is an important risk factor for neonatal death, morbidity and abnormal childhood development. Premature birth is currently regarded as a complex disease influenced by multiple factors. Common risk factors include nutritional deficiency during pregnancy, maternal obesity, environmental exposure, infection and inflammation, among which maternal nutrition during pregnancy is an important modifiable factor.

OBJECTIVE: To assess the relationship between maternal dietary calcium, phosphorus intake, and calcium supplement use before and during pregnancy was associated with the risk of PTB in offspring.

METHOD: This study was a nested case-control study conducted based on a large cohort study. And included pregnant women who were registered at the Perinatal Medicine Center of Gansu Provincial Maternal and Child Health Hospital from March 2018 to March 2019 and whose birth outcomes could be followed up. One-on-one dietary interviews were conducted during pregnancy, and a database was established based on the overall dietary intake levels for subsequent statistical analysis. PTB was defined as the outcome variable, while the intake levels of different substances during pregnancy were set as independent variables. Unconditional logistic regression models estimated the association between nutrient intake and the risk of PTB. Calculating the odds ratio (OR) and its 95% confidence interval (CI) to analyze the impact of different substance intake levels on PTB. Additionally, a restricted cubic spline (RCS) model with multivariable adjustment was applied to excess the non-linear association between dietary magnesium and calcium intake was associated with the risk of PTB.

RESULT: A total of 8897 pregnant women were included in the study, with 880 assigned to the case group and 8017 to the control group. Multivariate logistic regression analysis showed that low phosphorus intake in the second trimester was associated with an increased risk of PTB (OR = 1.297, CI: 1.020-1.649, P = 0.0341). Furthermore, similar results also exist for non-use of calcium supplements during the third trimester and low calcium intake preconception and during pregnancy. In addition, calcium, phosphorus and calcium supplements have a synergistic effect was associated with the risk of PTB.

CONCLUSION: During the second and third trimesters of pregnancy, the intake of phosphorus and the use of calcium supplements should be increased. Additionally, to prevent premature birth, the intake of calcium should be increased preconception and during pregnancy. Furthermore, this might lead to the optimization of public health policies or the formulation of guidelines for prenatal nutrition.

PMID:41484936 | DOI:10.1186/s41043-025-01211-8

Commun Chem. 2026 Jan 3. doi: 10.1038/s42004-025-01866-8. Online ahead of print.

ABSTRACT

Machine learning (ML) and artificial intelligence (AI) techniques are transforming the way chemical reactions are studied today. Datasets from high-throughput experimentation (HTE) are generated to better understand the reaction conditions crucial for outcomes such as yields and selectivities. However, it is often overlooked that datasets from such designed experiments possess a specific structure, which can be captured by a statistical model. Ignoring these data structures when applying ML/AI algorithms can result in misleading conclusions. In contrast, leveraging knowledge about the data-generating process yields reliable, interpretable, and comprehensive insights into reaction mechanisms. A particularly complex dataset is available for the Buchwald-Hartwig amination. Using this dataset, a statistical model for such HTE-generated chemical data is introduced, and a parameter estimation algorithm is developed. Based on the estimated model, new insights into the Buchwald-Hartwig amination are discussed. Our approach is applicable to a wide range of HTE-generated data for chemical reactions and beyond.

PMID:41484279 | DOI:10.1038/s42004-025-01866-8

Eye (Lond). 2026 Jan 4. doi: 10.1038/s41433-025-04229-8. Online ahead of print.

NO ABSTRACT

PMID:41484256 | DOI:10.1038/s41433-025-04229-8

Eur Radiol. 2026 Jan 4. doi: 10.1007/s00330-025-12272-y. Online ahead of print.

ABSTRACT

OBJECTIVES: To evaluate the effect of prostate volume on the risk of clinically significant prostate cancer (csPCa) across a range of PSA-density (PSAd) values, and to explore the relationship between PI-RADS category and PSAd in predicting csPCa.

MATERIALS AND METHODS: We retrospectively analyzed 2190 patients undergoing mpMRI for suspected PCa. Patients were classified as csPCa and clinically insignificant (negative and insignificant PCa). Logistic regression was performed to assess the csPCa risk at different PSAd cut-offs across different prostate volume subgroups (< 40, 40-60, 60-80, > 80 mL) and PI-RADS categories. The effect of prostate volume on PSAd performance was evaluated using ROC analysis. To assess robustness, we performed an 80:20 split-sample internal validation.

RESULTS: 747/2190 (34.1%) patients had PCa, including 571 (26.1%) with csPCa. Regardless of PSAd, csPCa risk exceeded 30% for PI-RADS 4 and 50% for PI-RADS 5. At a 10% csPCa risk threshold, the optimal PSAd cut-offs were 0.20 ng/mL² for PI-RADS 1-2 and 0.12 ng/mL² for PI-RADS 3. Logistic regression showed a significant inverse correlation between prostate volume and csPCa probability. Notably, 79% of csPCa patients with prostate volume ≤ 40 mL had a PSAd ≥ 0.15 ng/mL², compared to only 22.4% with volumes ≥ 60 mL. PSAd performed significantly worse for larger glands (≥ 60 mL), with AUCs of 0.70 versus 0.84 (≤ 40 mL) and 0.82 (40-60 mL), both p < 0.001.

CONCLUSION: The optimal PSAd cut-offs for guiding biopsy decisions were 0.20 ng/mL² for PI-RADS 1-2 and 0.12 ng/mL² for PI-RADS 3. When using PSAd to guide biopsy decision for PI-RADS 1-3 patients with large prostates (> 60 mL), caution is warranted, as PSAd becomes significantly less accurate.

KEY POINTS: Question The optimal PSA-density thresholds for biopsy decisions in PI-RADS 1-3 patients remain uncertain, and data on the impact of prostate volume on its performance are limited. Findings Optimal PSA-density thresholds were 0.20 ng/mL² for PI-RADS 1-2 and 0.12 ng/mL² for PI-RADS 3. Diagnostic performance of PSA-density decreased significantly in men with larger glands (≥ 60 mL). Clinical relevance PSA-density cut-offs of 0.20 (PI-RADS 1-2) and 0.12 ng/mL² (PI-RADS 3) can guide biopsy decisions. In PI-RADS 1-3 patients with large prostate (≥ 60 mL), PSA-density becomes significantly less predictive, and low values may not reliably exclude csPCa.

PMID:41484253 | DOI:10.1007/s00330-025-12272-y

Commun Eng. 2026 Jan 3. doi: 10.1038/s44172-025-00576-2. Online ahead of print.

ABSTRACT

Large-volume 3D dense prediction is essential in industrial applications like energy exploration and medical image segmentation. However, existing deep learning models struggle to process full-size volumetric inputs at inference due to memory constraints and inefficient operator execution. Conventional solutions-such as tiling or compression-often introduce artifacts, compromise spatial consistency, or require retraining. Here we present a retraining-free inference optimization framework that enables accurate, efficient, whole-volume prediction without performance degradation. Our approach integrates operator spatial tiling, operator fusion, normalization statistic aggregation, and on-demand feature recomputation to reduce memory usage and accelerate runtime. Validated across multiple seismic exploration models, our framework supports full size inference on volumes exceeding 1024³ voxels. On FaultSeg3D, for instance, it completes inference on a 1024³ volume in 7.5 seconds using just 27.6 GB of memory-compared to conventional inference, which can handle only 448³ inputs under the same budget, marking a 13 × increase in volume size without loss in performance. Unlike traditional patch-wise inference, our method preserves global structural coherence, making it particularly suited for tasks inherently incompatible with chunked processing, such as implicit geological structure estimation. This work offers a generalizable, engineering-friendly solution for deploying 3D models at scale across industrial domains.

PMID:41484251 | DOI:10.1038/s44172-025-00576-2

Sci Rep. 2026 Jan 3. doi: 10.1038/s41598-025-33777-y. Online ahead of print.

ABSTRACT

Volatile Organic Compounds (VOCs) from nail products pose potential health risks to technicians and clients. The present study investigates the exposure of nail technicians to VOC and assesses the effectiveness of dilution ventilation in mitigating these exposures. A test chamber was configured to simulate a nail salon environment, where three common activities were performed: applying nail polish, removing nail extensions, and nail extension application. A MultiRae Lite gas meter was used to monitor VOC concentrations in the technicians’ breathing zones and the ambient environment under both non-ventilated and dilution-ventilated conditions. The average VOC concentrations in the technicians’ breathing zones were measured at 8.3 (± 8.576) ppm, 116.2 (± 120.04) ppm, and 29.73 (± 7.876) ppm during nail polish application, nail extension removal, and nail extension application, respectively (n = 30 measured over 15 min). For the same activities, the average ambient VOC concentrations in the non-ventilated chamber were 25.03 (± 13.006) ppm, 192.17 (± 114.900) ppm, and 40.90 (± 30.891) ppm. These concentrations were significantly lowered by dilution ventilation to 0.93 (± 0.179) ppm, 12.6 (± 5.21) ppm, and 2.63 (± 0.812) ppm. For all three activities, a t-test verified a statistically significant drop in VOC concentration with dilution ventilation (p < 0.001). Additionally, the study discovered that VOC emissions could be reduced by over 65% at an air change rate (ACH) of 10 per hour. The results indicate that dilution ventilation can significantly decrease VOC exposure; nonetheless, the study concludes that it is insufficient as the sole control strategy in nail salons because concentrations failed to reach the limit levels, even at the highest tested ACH (100 ACH). This research provides a new scientific perspective on ventilation as a means of controlling occupational exposure to VOC in nail salons.

PMID:41484233 | DOI:10.1038/s41598-025-33777-y

Sci Rep. 2026 Jan 2. doi: 10.1038/s41598-025-34330-7. Online ahead of print.

ABSTRACT

Phytophthora rot, caused by the pathogen Phytophthora capsici, is one of the most destructive diseases affecting a wide range of plants, from vegetables to trees. This disease can attack the roots, aerial organs, and even the fruits of the affected plants. One of the latest control methods involves utilizing DNA interference (DNAi) technology, which employs short DNA oligonucleotides to disrupt the expression of pathogenic genes. In this research study, a 26-nucleotide double-stranded OligoDNA fragment, synthesized from the region of the cellulose synthase gene (CesA3), was used to regulate the expression of CesA3. The oligonucleotide sequence was encapsulated using gelatin and whey protein concentrate (WPC), separately, through electrospraying in an electrospinning machine. The physicochemical properties of the produced microcapsules were then evaluated. Pathogenicity tests were conducted using a completely randomized design, which involved baiting hemp seeds infected with P. capsici and treating the roots of bell pepper plants with the encapsulated OligoDNA. The development of the disease was monitored daily for two months, and the disease severity index was assessed. Statistical analysis, using the paired t-test, was performed on the collected data. The pathogen was cultured in a Corn meal agar (CMA) medium to investigate further the impact of the encapsulated OligoDNA on the growth of P. capsici. The results showed that OligoDNA encapsulated with the WPC polymer significantly reduced the disease index (DI) of P. capsici to approximately 16%, compared to other treatments, including the control, WPC, G, and G-Oligo, which exhibited DIs ranging from 35 to 40%. In the colony growth assay, both treatments, WPC and WPC-Oligo DNA, result in reduced growth compared to the control, with WPC-Oligo DNA having the most suppressive effect.

PMID:41484212 | DOI:10.1038/s41598-025-34330-7

Sci Rep. 2026 Jan 2. doi: 10.1038/s41598-025-34282-y. Online ahead of print.

ABSTRACT

Curcumin’s clinical utility is limited by poor bioavailability and dose-dependent toxicity. Although nano-encapsulation can address these shortcomings, rationally optimizing nanocarrier biosafety remains challenging due to the highly multidimensional design space. Here, we develop an interpretable Physics-Informed Machine Learning (PIML) framework that integrates experimental data from 75 curcumin nanocarriers with DLVO stability theory and drug-release kinetics to predict and optimize cytotoxicity. Among the evaluated models, XGBoost attained the highest statistical performance (R² = 0.89), although the PIML model provided physically coherent predictions with similar accuracy (R² = 0.86). SHAP research indicated a moderate negative zeta potential (-30 to -40 mV), chitosan-based coatings, and particle sizes of 150-250 nm as the principal factors contributing to decreased cytotoxicity. Multi-objective Bayesian optimization delineated a Pareto-optimal design space, facilitating approximately 82% toxicity reduction compared to free curcumin, while preserving around 70% loading efficiency. The study develops a proven, generalizable computational methodology that converts intricate nanocarrier design interactions into practical guidelines for the fabrication of safer curcumin-based nanotherapeutics.

PMID:41484189 | DOI:10.1038/s41598-025-34282-y

Sci Rep. 2026 Jan 3. doi: 10.1038/s41598-025-27634-1. Online ahead of print.

ABSTRACT

Reliable cause-specific mortality statistics are crucial for defining health priorities, public health programs, allocating resources, designing and implementing policies to improve healthcare quality and accessibility. India accounts for almost 18 percent of the world’s population. The 2020 report from the Office of the Registrar General of India indicates that the Medical Certification of Cause of Death (MCCD) rate is only 22.5%, with a minimal improvement of just 2.5% over the past decade. This study is the first to provide a comprehensive evaluation of MCCD-patterns across India over the past 15 years, addressing a critical-gap in the literature by identifying regional patterns, disparities, and healthcare variables that have previously been underexplored. Based on MCCD-trends over this period, the states and Union-Territories of India can be categorized into three clusters. Cluster1 includes 23 states with the lowest-average MCCD-rate of 18%, attributed to a low 0.14 doctors per 1000 people, with only 27.4% of hospitals actively reporting-MCCD. In contrast, Clusters2 and 3 have higher-average MCCD-rates of 63% and 60%, respectively, supported by higher 0.27 and 0.33 doctors per 1000 people, with over 80% of hospitals actively reporting-MCCD. Although, the findings indicate that active MCCD-reporting is a major factor associated with MCCD rates, other factors including healthcare infrastructure, state-specific healthcare policies, socioeconomic factors, and administrative management also influence MCCD-rates.

PMID:41484187 | DOI:10.1038/s41598-025-27634-1