Tag: statistics

Fam Pract. 2025 Aug 14;42(5):cmaf071. doi: 10.1093/fampra/cmaf071.

ABSTRACT

BACKGROUND: The aim of the study was to assess the effect of an interdisciplinary, digital consultation platform on the proportion of appropriate referrals from general practitioners (GPs) to an orthopaedic outpatient hospital.

METHODS: We performed a stepped wedge, cluster, randomized controlled trial. Sixty GP practices in the catchment area of a large teaching hospital in the Netherlands were randomized. Groups of GP practices entered the trial in four steps at 13-week intervals, at which point they received access to the Prisma platform. The platform allowed them to post questions about anonymized cases to a multidisciplinary group of specialists. During the control condition, GPs did not receive platform access. In both conditions, GPs provided care as usual. The proportion of appropriate referrals, defined as referrals for which a patient had either (i) more than one consultation with an orthopaedic surgeon or (ii) one consultation with additional diagnostics or interventions, was the primary outcome.

RESULTS: Participating GPs referred 4928 patients to hospital. Intention-to-treat analysis showed a 4.4% estimated increase in the proportion of appropriate referrals among GP practices randomized to have access to the platform compared to the control group, with an odds ratio (OR) of 1.22 [95% confidence interval (CI), 1.01-1.46; P = 0.037]. Per-protocol analysis showed a smaller, but non-significant, 2.2% difference between interventions, with an OR of 1.11 (95% CI, of 0.96%-1.28%; P = 0.178).

CONCLUSIONS: We observed a modest increase in appropriate referrals for orthopaedic review among GP practices randomized to the platform. On a larger scale, this could contribute to more sustainable access to specialist care.

PMID:40973675 | DOI:10.1093/fampra/cmaf071

Med Phys. 2025 Oct;52(10):e70019. doi: 10.1002/mp.70019.

ABSTRACT

BACKGROUND: Prostate cancer (PCa) presents a significant global health challenge affecting men. Accurate segmentation and grading of PCa lesions in multiparametric Magnetic Resonance Imaging (mp-MRI) are essential for effective diagnosis and treatment planning.

PURPOSE: This study aimed to develop and validate an automated model for PCa lesion segmentation and Prostate Imaging Reporting and Data System (PI-RADS) grading in mp-MRI.

METHODS: The lesion’s perceived characteristics are strongly related to both imaging modalities and lesion locations. Therefore, we propose a Lesion-guided Selective Multi-modal Integration (LeSMI) module. This module incorporates two advanced mechanisms-Dynamic Modality Weighting (DMW) and Localized Lesion Attention (LLA)-to dynamically integrate crucial information across and within imaging modalities. Specifically, DMW operates on the mp-MRI inputs (T2-weighted (T2w) images, diffusion-weighted imaging (DWI), and apparent diffusion coefficient (ADC) maps) to dynamically assign weights to each modality, thereby integrating complementary information and enhancing feature identification across different contexts. LLA, on the other hand, maintains spatial structure information within each modality for precise lesion localization. Inspired by clinical workflows, our framework is employed through a two-stage Prostate Cancer Segmentation and Grading (PCaSG) strategy, leveraging knowledge from segmentation to improve PI-RADS grading performance. We validated our method using two publicly available datasets, namely, Prostate158 and PI-CAI Challenge, to assess its advantages over other methods. For the Prostate158 dataset, we used the officially reported partition with 119 cases for training, 20 for validation, and 19 for testing. In contrast, the PI-CAI Challenge dataset, which lacks predefined splits, was randomly divided into 180 for training, 20 for validation, and 20 for testing. In addition to these dataset partitions, 5-fold cross-validation was conducted on both the Prostate158 and PI-CAI Challenge datasets to provide a more robust and comprehensive statistical evaluation of the model’s performance.

RESULTS: Evaluated on the Prostate158 and PI-CAI Challenge datasets, our method demonstrated superior performance, achieving a Dice Similarity Coefficient (DSC) of 51.30% and a lesion-level quadratic-weighted kappa score ( Q W K l $QW{{K}_l}$ ) of 62.48% on Prostate158, and a DSC of 43.81% and a Q W K l $QW{{K}_l}$ of 42.98% on PI-CAI. These results represent improvements of up to 2% in DSC and 17% in Q W K l $QW{{K}_l}$ over current state-of-the-art models on Prostate158, and enhancements of 4% in DSC and 3% in Q W K l $QW{{K}_l}$ on PI-CAI.

CONCLUSION: The proposed model’s robustness in handling diverse lesion presentations, combined with its reliable assessments, underscores its significant clinical applicability. Our model offers substantial advancements in both segmentation accuracy and PI-RADS grading, addressing the challenges of inter-reader variability and the need for high expertise in conventional diagnostic practices. This technological innovation holds promise for enhancing early, accurate detection and risk assessment in prostate cancer management, ultimately improving patient outcomes.

PMID:40973673 | DOI:10.1002/mp.70019

Bioinformatics. 2025 Sep 19:btaf520. doi: 10.1093/bioinformatics/btaf520. Online ahead of print.

ABSTRACT

MOTIVATION: The gut microbiome plays an important role in human health and disease, prompting large-scale studies that generate extensive datasets. A critical preprocessing step in analyzing such datasets is anomaly detection, which aims to identify erroneous samples and prevent misleading statistical outcomes. Microbiome data, however, pose unique challenges such as compositionality, sparsity, interdependencies, and high dimensionality, limiting the effectiveness of conventional methods and highlighting the need for specifically-tailored approaches for anomaly detection in microbiome data.

IMPLEMENTATION: To address this challenge, we introduce KADAIF, a microbiome-specific anomaly detection method that generalizes the common Isolation Forest approach. As in Isolation Forest, KADAIF builds an ensemble of trees, each recursively partitioning the data along randomly selected features, and measures the average depth at which samples are isolated, assuming that anomalous samples will be isolated closer to the root. Unlike Isolation Forest, however, KADAIF partitions samples based on subsets of features (coupled with dimensionality reduction), addressing microbiome-specific properties such as sparsity and species interactions.

RESULTS: We evaluate KADAIF by simulating common scenarios that introduce anomalous behavior, demonstrating that KADAIF outperforms alternative methods across various settings and datasets. Furthermore, we show that KADAIF outperforms Isolation Forest in detecting anomalies also in other types of high dimensional sparse biological data. Finally, we show KADAIF’s application for identifying disease onset in longitudinal microbiome data and for partitioning cases vs controls based on the Anna Karenina principle. Combined, our work highlights KADAIF’s potential to enhance microbiome data processing and downstream analyses, with beneficial implications for precision medicine studies.

AVAILABILITY: An implementation of KADAIF, as well as all the code used for the analysis, is available on GitHub (https://github.com/borenstein-lab/KADAIF).

SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

PMID:40973672 | DOI:10.1093/bioinformatics/btaf520

J Occup Health. 2025 Sep 19:uiaf053. doi: 10.1093/joccuh/uiaf053. Online ahead of print.

ABSTRACT

OBJECTIVES: This study evaluated the impact of transitioning from an 8-hour to a 12-hour shift schedule on sleep patterns and well-being in intensive care unit (ICU) nurses with preexisting sleep disturbances using wearable sensors. We also examined differences in outcome based on chronotype.

METHODS: We conducted an observational study at a university hospital ICU between November 2020 and October 2023, before and after a hospital-wide shift schedule change. Nurses wore wearable sensors and completed daily surveys over five weeks under each shift system. Rotating-shift ICU nurses with a Pittsburgh Sleep Quality Index (PSQI) score>5 were eligible. Sleep metrics and subjective well-being were compared using linear mixed models, adjusting for age. Sleep episodes were categorized relative to shift timing, and chronotype-stratified subgroup analyses were performed.

RESULTS: Eighty nurses completed the study (12-h shift: 37; 8-h shift: 43). The interval between shifts was greater for the 12-h shift group (36.12 vs 26.78 hours). Total sleep duration did not significantly differ between groups(12-h shift: 418.5 minutes; 8-h shift: 398 minutes); however, the 12-h shift group had less fragmented sleep, higher subjective well-being scores, and lower reported stress and fatigue. Evening chronotypes appeared to benefit more from 12-h shifts, with longer sleep duration and higher well-being scores, though these differences were not statistically significant.

CONCLUSIONS: Transitioning to a 12-hour shift schedule was associated with reduced sleep fragmentation and improved well-being, particularly among evening chronotypes. These findings suggest that shift schedule structure and individual chronotype may influence adaptation to shift work in ICU settings.

PMID:40973662 | DOI:10.1093/joccuh/uiaf053

Sleep. 2025 Sep 19:zsaf282. doi: 10.1093/sleep/zsaf282. Online ahead of print.

ABSTRACT

STUDY OBJECTIVES: GGIR is an R package for processing raw acceleration data to estimate sleep health parameters. We aimed to 1) assess the performance of three sleep algorithms within GGIR against PSG for detecting sleep/wake in clinically referred, typically-developing children (criterion validity); and 2) describe GGIR-derived sleep estimates from typically developing children enrolled in multiple cohort studies (face validity).

METHODS: For criterion evaluation, children (8-16y, N=30) wore an actigraphy device for one night during in-lab polysomnography with performance assessed using epoch-by-epoch analyses. For face validity evaluation, four community/free living datasets were used: 1) BMAYC (3-5y, N=310), 2) SSS (5-8y, N=118), 3) S-Grow2 (12-13y; N=291) and 4) ELEMENT (9-18y; N=543). All raw acceleration data were processed using GGIR (v.3.0-0) with the Cole-Kripke (CK), Sadeh (S), and van Hees (vH) algorithm settings.

RESULTS: Following the in-lab test, 60% of children were diagnosed with mild to severe obstructive sleep apnea (OSA). For criterion evaluation, the 30-s epoch-by-epoch analyses revealed that average balanced accuracies were 0.80 (Sensitivity=0.80; Specificity=0.79), 0.76 (Sensitivity=0.86; Specificity=0.65), and 0.67 (Sensitivity=0.95, Specificity=0.39) for GGIR-CK, GGIR-vH, and GGIR-S, respectively. For face validity evaluation, sleep estimates mirrored the in-lab performance metrics (e.g., sleep duration estimates were similar when using GGIR-CK and GGIR-VH but approximately one hour longer when using GGIR-S).

CONCLUSIONS: The in-lab performance metrics, from typically-developing children with and without OSA, and cohort-based descriptive statistics from samples of typically-developing children, provide benchmark data to guide investigators on the suitability of GGIR for automated processing of raw acceleration data for pediatric sleep estimation.

PMID:40973655 | DOI:10.1093/sleep/zsaf282

J Burn Care Res. 2025 Sep 19:iraf181. doi: 10.1093/jbcr/iraf181. Online ahead of print.

ABSTRACT

BACKGROUND: ReCell, an autologous cell harvesting technology also known as Autologous Skin Cell Suspension (ASCS), has shown promise in enhancing skin regeneration for burn patients. Despite its growing use in clinical practice, the current literature displays considerable variability in study design and quality, leading to ongoing uncertainty about its true clinical effectiveness. This systematic review and meta-analysis aim to comprehensively evaluate the clinical efficacy of ASCS in the treatment of burns.

METHODS: A systematic review was conducted in accordance with PRISMA guidelines, utilizing searches across PubMed, Web of Science, Embase, and Cochrane databases. The review protocol was prospectively registered on PROSPERO (CRD42024606554). The Cochrane Risk of Bias Tool and the ROBINS-I tool were applied to assess bias in randomized controlled trials and observational studies, respectively. The overall methodological quality of included studies was appraised using the GRADE framework.

RESULTS: Fourteen studies (n = 3362) fulfilled the inclusion criteria. The pooled mean patient age was 37.6 years, with a male predominance (65.9%). The average %TBSA affected was 14.6% (95% CI: 8.8-20.4), with substantial heterogeneity (I2 = 95.1%). Meta-analysis demonstrated a statistically significant reduction in complication rates with ASCS combined with split-thickness skin grafting (STSG) compared to STSG alone (RR = 0.64, 95% CI: 0.41-1.00, p = 0.048). However, rates of wound infection and graft failure did not differ significantly between groups.

CONCLUSIONS: ASCS demonstrates potential in reducing complications in burn care. Nevertheless, due to heterogeneous study designs, further high-quality, large-scale randomized trials are warranted to validate its long-term efficacy and broader clinical utility.

PMID:40973648 | DOI:10.1093/jbcr/iraf181

Genome Biol Evol. 2025 Sep 19:evaf177. doi: 10.1093/gbe/evaf177. Online ahead of print.

ABSTRACT

Deep learning (DL) has been widely used in various scientific fields, but its integration into phylogenetics has been slower, primarily due to the complex nature of phylogenetic data. The studies that apply DL to sequencing data often limit analyses to 4-taxon trees. Many of these studies serve as “proof of principle” and perform similarly to traditional phylogeny reconstruction methods. New ways of using training data, such as encoding with compact bijective ladderized vectors or transformers, enable the handling of much larger trees and genomic data sets. This short perspective focuses on the application of DL in phylogenetics, introducing prevalent DL architectures. We highlight potential problems in the field by discussing the risks of using simulation-based training data and emphasize the importance of reproducibility and robustness in computational estimates. Finally, we explore promising research areas, including the combination of phylogenetics and population genetics in DL, the analysis of neighbor dependencies, and the potential to significantly reduce computational cost compared to traditional methods. This perspective illustrates the potential of DL in complementing traditional phylogeny reconstruction methods and aiding the advancement of phylogenetic analysis, especially in performing computationally demanding tasks such as model selection or estimating branch support values.

PMID:40973626 | DOI:10.1093/gbe/evaf177

Trans R Soc Trop Med Hyg. 2025 Sep 19:traf104. doi: 10.1093/trstmh/traf104. Online ahead of print.

ABSTRACT

As part of a regional initiative to enhance epidemic preparedness, the Dominican Republic became one of 16 national hubs in the Artificial Intelligence for Pandemic and Epidemic Preparedness and Response Network. This article outlines the early steps taken to introduce responsible artificial intelligence into public health practice by developing local capacity, predictive surveillance tools and ethical governance strategies. Drawing on implementation experiences within the country’s health institutions and communities, this article highlights practical lessons and operational insights. These findings support broader discussions on equity-focused digital innovation and provide a replicable model for low- and middle-income countries seeking to strengthen their readiness for future health threats.

PMID:40973619 | DOI:10.1093/trstmh/traf104

Clin Lung Cancer. 2025 Aug 26:S1525-7304(25)00219-0. doi: 10.1016/j.cllc.2025.08.012. Online ahead of print.

ABSTRACT

BACKGROUND: In treatment-naïve advanced non-small cell lung cancer (NSCLC) patients with epidermal growth factor receptor (EGFR) mutations, approximately 10% to 15% correspond to uncommon mutations. For patients with EGFR uncommon mutations, monotherapy with either a second-generation EGFR tyrosine kinase inhibitor (TKI), afatinib, or a third-generation EGFR-TKI, osimertinib, is recommended as the initial therapy. However, needs remain unmet for patients with central nervous system (CNS) metastases and those who do not respond adequately to single-agent TKI therapy for EGFR uncommon mutations. The recently published FLAURA2 trial showed that osimertinib in combination with platinum-pemetrexed significantly prolonged progression-free survival (PFS) and provided high disease control compared with osimertinib monotherapy for common mutations. Therefore, we planned this phase II study to evaluate the efficacy and safety of osimertinib in combination with platinum-pemetrexed in treatment-naïve NSCLC patients with EGFR uncommon mutations.

PATIENTS AND METHODS: Forty patients will be enrolled in the study. The primary endpoint is the objective response rate, and the secondary endpoints include safety, PFS and overall survival in overall patients, patients with and without CNS lesions at baseline and according to mutation subtype.

CONCLUSIONS: In this study, we will explore the efficacy and safety of osimertinib in combination with platinum-pemetrexed in treatment-naïve NSCLC patients with EGFR uncommon mutations. Our findings may provide treatment options for patients with EGFR uncommon mutations, especially those with CNS metastases or those requiring more intensive treatment.

PMID:40973606 | DOI:10.1016/j.cllc.2025.08.012

Clin Ther. 2025 Sep 18:S0149-2918(25)00271-1. doi: 10.1016/j.clinthera.2025.08.001. Online ahead of print.

ABSTRACT

PURPOSE: Tumor lysis syndrome (TLS) is a life-threatening metabolic emergency caused by rapid tumor cell breakdown, either spontaneously or after therapy, leading to electrolyte imbalances that can result in acute kidney injury, arrhythmias, seizures, and multiorgan failure. Despite its clinical importance, the relationship between anticancer drugs and TLS, particularly newer targeted therapies, remains poorly understood.

METHODS: We analyzed the US Food and Drug Administration (FDA) Adverse Events Reporting System database, a repository of adverse events associated with medical products, to identify TLS cases reported from the first quarter of 2004 to the third quarter of 2024. For signal detection, we used disproportionality analysis with 4 algorithms-reported odds ratio, proportional reporting ratio, Bayesian confidence propagation neural network, and empirical Bayes geometric mean. These algorithms assessed statistical correlations between anticancer drugs and TLS, based on a 2 × 2 contingency table framework.

FINDINGS: From the first quarter of 2004 to the third quarter of 2024, a total of 7340 TLS cases were documented in the FDA Adverse Events Reporting System database. Clinical characteristics, including age, sex, and outcomes, were analyzed. Among all reported TLS cases, 53.0% were men, and the mean age across all individuals was 56.9 ± 21.5 years. The incidence of TLS peaked in 2022, with a 42% increase from 2016 to 2017. A total of 118 antineoplastic drugs were identified as highly associated with TLS, of which only 18 had FDA-labeled TLS-related adverse reactions. Chemotherapy drugs were the most frequently associated with TLS. Venetoclax emerged as the top drug associated with TLS, comprising 10.72% of all TLS reports.

IMPLICATIONS: Our findings highlight critical drug-induced TLS associations, particularly with emerging targeted therapies such as venetoclax. The study underscores the need for clinicians to monitor TLS closely in patients receiving certain anticancer treatments and to refine therapeutic strategies to mitigate TLS risk, ensuring safer cancer care outcomes. Further longitudinal studies are warranted to validate these findings and enhance pharmacovigilance efforts.

PMID:40973598 | DOI:10.1016/j.clinthera.2025.08.001