Category: Nevin Manimala

J Imaging Inform Med. 2025 Dec 10. doi: 10.1007/s10278-025-01752-8. Online ahead of print.

ABSTRACT

To reduce the amount of required labeled data for lung disease severity classification from chest X-rays (CXRs) under class imbalance, this study applied deep active learning with a Bayesian Neural Network (BNN) approximation and weighted loss function. This retrospective study collected 2319 CXRs from 963 patients (mean age, 59.2 ± 16.6 years; 481 females) at Emory Healthcare affiliated hospitals between January and November 2020. All patients had clinically confirmed COVID-19. Each CXR was independently labeled by 3 to 6 board-certified radiologists as normal, moderate, or severe. A deep neural network with Monte Carlo Dropout was trained using active learning to classify disease severity. Various acquisition functions were used to iteratively select the most informative samples from an unlabeled pool. Performance was evaluated using accuracy, area under the receiver operating characteristic curve (AU-ROC), and area under the precision-recall curve (AU-PRC). Training time and acquisition time were recorded. Statistical analysis included descriptive metrics and performance comparisons across acquisition strategies. Least Confidence achieved 92.8% accuracy (AU-ROC, 0.95) in binary classification (normal vs. diseased) using 9.24% of the training data. In the multi-class setting, Mean STD achieved 70.5% accuracy (AU-ROC, 0.85) using 21.87% of the labeled data. These methods outperformed more complex and computationally expensive acquisition functions and significantly reduced labeling needs. Deep active learning with BNN approximation and weighted loss effectively reduces labeled data requirements while addressing class imbalance, maintaining or exceeding diagnostic performance.

PMID:41369959 | DOI:10.1007/s10278-025-01752-8

JAMA Dermatol. 2025 Dec 10. doi: 10.1001/jamadermatol.2025.4929. Online ahead of print.

ABSTRACT

IMPORTANCE: Acute digital ischemia, digital ulcers, and gangrene are debilitating complications of systemic sclerosis and other vasculopathies and are often refractory to standard vasodilator and immunosuppressive therapies. Botulinum toxin (BTX) has emerged as a potential rescue therapy, but its clinical effectiveness and safety remain unclear.

OBJECTIVE: To evaluate the effectiveness and safety of BTX injections for ischemic digital complications and identify predictors of treatment response using individual participant data (IPD).

DATA SOURCES: MEDLINE (PubMed), Embase (Ovid), and Scopus were searched from inception through April 20, 2024.

STUDY SELECTION: Eligible studies included patients who presented with acute digital ischemia, ischemic digital ulcers, or gangrene. Studies were limited to Raynaud disease without digital ulcers or gangrene were excluded. Two reviewers independently screened articles using Covidence, with discrepancies resolved by consensus with the senior author. Of 116 studies screened, 31 (27%) met inclusion criteria.

DATA EXTRACTION AND SYNTHESIS: Data were extracted in duplicate and study quality was assessed using the Joanna Briggs Institute checklist. Descriptive statistics were used to summarize baseline characteristics, treatment regimens, and outcomes.

MAIN OUTCOMES AND MEASURES: The primary outcome was complete response (CR), which was defined as resolution of ischemia or ulcer healing. Secondary outcomes included adverse events and time to response. Cox regression was used to identify factors associated with CR.

RESULTS: This systematic review and IPD meta-analysis included 119 patients (72 female individuals [75.0%]; mean [SD] age, 49.0 [15.1] years). BTX was associated with high CR rates for ischemia (93.1%), ulcers (90.1%), and gangrene (87.5%). Adverse events were infrequent, with transient muscle weakness (7.6%) and injection site pain (5.9%) being most common. No associated factors reached statistical significance in multivariable models, but autoimmune etiology and younger age were associated with faster response in Kaplan-Meier analyses.

CONCLUSIONS AND RELEVANCE: The results of this systematic review and IPD meta-analysis suggest that BTX injections appear to be a safe and effective adjunct for refractory digital ischemia in systemic sclerosis. Prospective trials are needed to confirm long-term effectiveness and standardize administration protocols.

PMID:41369954 | DOI:10.1001/jamadermatol.2025.4929

Pharm Stat. 2026 Jan-Feb;25(1):e70051. doi: 10.1002/pst.70051.

ABSTRACT

In clinical research, it is increasingly difficult to conduct fully powered and well-balanced randomized controlled trials, particularly when studying rare or devastating diseases and pediatric patients. While Bayesian methodologies are very useful for leveraging historical control data to meet some of these challenges, many practical and statistical concerns emerge when prospectively specifying a design to implement Bayesian methods. In this article, we discuss these concerns and propose novel methods to ensure statistical rigor when applying Bayesian methodology. A novel adaptive Bayesian borrowing (ABB) method proposed here borrows from historical control data to increase the precision of the control arm based on the observed congruence of the historical and current data. The method would also enable an adaptive increase of sample size to accommodate accumulating information. We demonstrate that this approach can be prospectively specified and provides a statistically rigorous and transparent inference while mitigating the risk of potential conflict between historical and current control data, as well as misspecifications of variability in the endpoints.

PMID:41369947 | DOI:10.1002/pst.70051

Clin Appl Thromb Hemost. 2025 Jan-Dec;31:10760296251401588. doi: 10.1177/10760296251401588. Epub 2025 Dec 10.

ABSTRACT

BackgroundThere is conflicting evidence of the correlation between fatty acid binding protein 4 (FABP4) and cardiovascular disease (CVD) in previous observational studies.ObjectiveThis study aims to explore the genetic causal association between FABP4 and adverse cardiovascular (CV) events.MethodsA two-sample bidirectional Mendelian randomization (MR) analysis was performed using summary statistics from GWAS. The primary method used for MR analysis was the inverse-variance weighted (IVW) method, complemented by MR-Egger, weighted median, and weighted mode methods to explore the causal association between FABP4 and adverse CV events. For sensitivity analysis assessing heterogeneity and pleiotropy, MR-Egger and MR-PRESSO were employed to address horizontal pleiotropy, while Cochran’s Q test was used to assess heterogeneity between instrumental variables (IVs). The leave-one-out analysis was used to detect outliers.ResultsIVW suggested that FABP4 showed no genetic causal effect on stroke (OR = 1.01, 95% CI = 0.98-1.04, p = 0.52), angina (OR = 1.0003, 95% CI = 0.9994-1.0012, p = 0.53), arrhythmia (OR = 1.0004, 95% CI = 0.9998-1.0009, p = 0.25), heart failure (OR = 0.99, 95% CI = 0.96-1.02, p = 0.53) or myocardial infarction (OR = 0.99, 95% CI = 0.97-1.01, p = 0.21). In the reverse MR analysis, IVW analysis showed no genetic causal effect of adverse CV events on FABP4. The results of other methods were consistent with the IVW method. Sensitivity analysis indicated the results was robust.ConclusionOur study did not find evidence to support a causal relationship between FABP4 and adverse CV events. Further studies are needed to comprehensively assess this potential association.

PMID:41369907 | DOI:10.1177/10760296251401588

Ann Hematol. 2025 Dec 10. doi: 10.1007/s00277-025-06598-2. Online ahead of print.

ABSTRACT

Quantification of foetal haemoglobin (HbF) via high-performance liquid chromatography (HPLC) remains a critical parameter for monitoring therapeutic response to hydroxyurea (HU) in patients with sickle cell disease (SCD). However, its routine application is constrained in low-income settings such as Nigeria due to limited accessibility. The platelet-to-neutrophil ratio (PNR), an emerging inflammatory biomarker, has demonstrated prognostic relevance in several disease conditions. This study investigates the potential utility of PNR as a surrogate monitoring tool in SCD patients receiving HU therapy. This was a retrospective analysis of data derived from the Children’s National Hospital Natural History Study of Sickle Cell Disease, conducted in Washington, D.C., USA. Medical records of patients with SCD were reviewed from 1 September 2013 to 14 September 2023. Associations between PNR, HU use, HbF concentration, and reticulocyte count were evaluated. T-tests was used for continuous variables and chi-square tests for categorical variables. Regression analysis was used to evaluate the predictors of hydroxyurea use. A p-value of < 0.05 was considered statistically significant. Of 437 patient records reviewed, 222 (50.8%) were male, and 312 (71.4%) were receiving HU. The majority (81%) had no history of transfusion. Patients on HU exhibited significantly higher platelet counts, HbF levels, and PNR values (p = 0.0008, p < 0.0001, and p < 0.0001, respectively). In multivariate logistic regression, both HbF (OR: 1.121; 95% CI: 1.051-1.196; p < 0.001) and PNR (OR: 1.014; 95% CI: 1.005-1.024; p = 0.0022) were independently associated with HU use. No significant associations were observed between HU use and age group, haemoglobin phenotype, transfusion history, or BMI (p > 0.05). The PNR demonstrates a significant association with HU use and HbF levels, indicating its potential as a low-cost, accessible biomarker for monitoring HU therapy in SCD. Although this analysis was based on data from a high-resource setting, the findings underscore the relevance of PNR as a feasible monitoring alternative in resource-constrained environments. Further validation in low-income contexts is warranted to establish population-specific reference intervals and inform clinical practice.

PMID:41369905 | DOI:10.1007/s00277-025-06598-2

Photochem Photobiol Sci. 2025 Dec 10. doi: 10.1007/s43630-025-00824-6. Online ahead of print.

ABSTRACT

PURPOSE: Temporomandibular disorder (TMD) pain originates from muscular or intracapsular disorders, with the latter being represented by arthralgia. The aim of this study was to investigate the effects of photobiomodulation therapy (PBMT) on nociceptor activity, proinflammatory cytokine expression, neuropeptide expression, and tissue alterations in temporomandibular joint (TMJ) discs following unilateral injury in rats.

METHODS: Disc injury was induced via surgical access to the TMJ under general anesthesia. Forty rats were divided into four groups (n = 10 each). Group 1: surgical injury to the articular disc with PBMT; Group 2: sham surgery with PBMT; Group 3: surgical injury to the articular disc without PBMT; and Group 4: naïve (control). Ten PBMT sessions were conducted with a GaAs laser at a wavelength of 904 nm and an energy density of 6 J/cm². TMJs were analyzed for histological and histomorphometric parameters to evaluate tissue changes and protein levels of substance P (SP), transient receptor potential vanilloid 1 (TRPV-1), calcitonin gene-related peptide (CGRP), interleukin-6 (IL-6), interleukin-1 beta (IL-1β), and tumor necrosis factor-alpha (TNF-α). Statistical analysis was performed by ANOVA with Tukey’s post-hoc test (p < 0.050 indicating a significant difference).

RESULTS: The results revealed increased expression of SP, TRPV-1, CGRP, IL-6, IL-1β, and TNF-α following TMJ injury, with significantly lower levels after PBMT than in the other groups, leading to an improvement in the initial phases of tissue repair.

CONCLUSION: These findings suggest that PBMT effectively modulates nociceptive activity and reduces proinflammatory cytokine expression, optimizing tissue regeneration and improving the treatment of TMD.

PMID:41369880 | DOI:10.1007/s43630-025-00824-6

Psychon Bull Rev. 2025 Dec 10;33(1):11. doi: 10.3758/s13423-025-02786-3.

ABSTRACT

The present study explores the effect of exposure to news in a social media environment on people’s perceived knowledge and on the “Illusion of knowledge” -that is, the overestimation of perceived knowledge relative to actual knowledge. Using a mixed, within-subjects design, participants (N = 828) engaged in a two-session study featuring pre- and post-exposure assessments. Participants scrolled through a social media newsfeed and completed assessments of perceived and actual knowledge. We hypothesized that (1) social media exposure would increase perceived knowledge, (2) this increase would not align with actual knowledge, and (3) higher topic involvement would exacerbate these effects. While most of the preregistered hypotheses did not receive clear statistical support, some effects were consistent with our predictions. Perceived knowledge increased over time, suggesting a general exposure effect; however, differences between exposed and non-exposed topics were not statistically significant, possibly due to test effects or limited engagement with the platform. A strong illusion of knowledge was observed across topics, suggesting robust overestimation. This high baseline may have created a ceiling that constrained the detection of exposure effects. This study was conducted as a Registered Report ( https://doi.org/10.24072/pci.rr.100986 ), accepted in principle prior to data collection. To our knowledge, it represents one of the most ecologically valid attempts in the literature to simulate social media exposure in an experimental setting.

PMID:41369854 | DOI:10.3758/s13423-025-02786-3

Discov Oncol. 2025 Dec 10. doi: 10.1007/s12672-025-04187-3. Online ahead of print.

ABSTRACT

BACKGROUND: Several studies have explored whether tobacco consumption is associated with the histologic growth patterns of pulmonary adenocarcinomas, and whether these patterns influence prognosis. However, findings have remained ambiguous. This study aimed at further exploring these associations in a larger cohort, enhancing precision compared to previous studies by assessing growth patterns on digitalized slides.

METHODS: Resection specimens from 196 patients with pulmonary adenocarcinoma were retrospectively assessed and categorized. Where available, up to three tumour bearing H&E stained slides were digitized. Growth patterns were quantitatively assessed and categorized based on these digital images. Clinical and demographic data, including smoking history, were obtained from patient admission records.

RESULTS: Smoking, particularly heavy smoking (> 30 pack-years), was significantly associated with solid growth, while non- or light smokers were more likely to display an acinar predominance. These two patterns were by far the most prevalent, together accounting for 84.3% of all cases. Median overall survival (OS) for the entire cohort was 47 months. Solid predominant tumours were associated with shorter OS (25 versus 57 months compared to all other predominant patterns, p = 0.042) while lepidic tumours were associated with longer OS (112 versus 46 months compared to all other predominant patterns, p = 0.025). No statistically significant correlation with OS was observed for either acinar, cribriform, or micropapillary tumours.

CONCLUSION: These findings support an association between smoking behaviour and histologic growth patterns, as well as between growth patterns and OS. The associations appear particularly robust for solid predominant tumour, highlighting their prognostic relevance.

PMID:41369828 | DOI:10.1007/s12672-025-04187-3

Mol Neurobiol. 2025 Dec 10;63(1):280. doi: 10.1007/s12035-025-05561-6.

ABSTRACT

ASD etiology may be influenced by non-coding single-nucleotide polymorphisms (SNPs) in 3′ untranslated regions (3’UTRs). These variations can affect post-transcriptional regulation by altering RNA structure and miRNA binding patterns. PTEN and TSC1, two key regulators of the PI3K/AKT/mTOR signaling pathway, are promising candidate genes for ASD. We performed a case-control study involving 108 individuals diagnosed with ASD and 184 healthy matched controls from the Tunisian population. Four specific 3’UTR SNPs (PTEN: rs701848, rs34140758; TSC1: rs739442, rs2809244) were analyzed through genotyping. Statistical associations were assessed using various genetic inheritance models, with multivariate logistic regression adjusting for gender, family psychiatric history, and parental age. A thorough bioinformatics approach was applied, which included miRNASNP-v3 for predicting altered miRNA binding sites, RNAhybrid for calculating the minimum free energy (MFE) of miRNA-mRNA duplexes, PhyloP and phastCons for evolutionary conservation analysis, and mfold for modeling RNA secondary structures. PTEN rs701848 showed a significant statistical association with ASD risk that remained significant after correction for multiple testing (corrected p < .0004), while TSC1 rs739442 showed a suggestive association (p = .009, corrected p = .036). Systematic functional assessment using multi-criteria evaluation identified rs701848 as having high predicted functional impact (extensive miRNA binding changes, including loss of neurodevelopmental miR-129 sites, high conservation (PhyloP = 2.24), and altered RNA stability), while rs739442 showed moderate impact, and rs2809244 minimal predicted functional significance. For mutant alleles, RNA structure modeling showed slight decreases in mRNA stability, which might affect transcript accessibility to regulatory factors. Evolutionary conservation analysis demonstrated that rs701848 and rs34140758 reside in highly conserved areas, whereas rs739442 and rs2809244 are in less evolutionarily conserved regions. By integrating population-genetics data with bioinformatics predictions, this study supports the relevance of non-coding SNPs as biological risk factors for ASD. It emphasizes their functional significance in modulating post-transcriptional gene regulation mechanisms.

PMID:41369821 | DOI:10.1007/s12035-025-05561-6

Ann Surg Oncol. 2025 Dec 10. doi: 10.1245/s10434-025-18862-w. Online ahead of print.

ABSTRACT

BACKGROUND: Lung cancer, especially non-small cell lung cancer (NSCLC), is a leading cause of cancer death. Surgical resection is the primary treatment for early-stage NSCLC. Video-assisted thoracoscopic surgery (VATS) and robot-assisted thoracic surgery (RATS) are minimally invasive options, with RATS offering better precision at higher costs. This meta-analysis compared the perioperative outcomes of both techniques in NSCLC.

METHODS: A systematic search in PubMed, Web of Science, NLM, CNKI, and Wanfang was conducted for RCTs published between 31 July 2021 and 31 December 2024. Primary outcomes included operation time, blood loss, lymph node dissection, conversion to thoracotomy, chest drainage time, and hospital stay. Statistical analyses were performed using RevMan 5.3, with heterogeneity assessed using the I² statistic. A p value lower than 0.05 was considered significant.

RESULTS: Three RCTs were analyzed. Blood loss showed no significant difference between VATs and RATs (mean difference [MD], -33.00; 95% CI, -85.51 to 22.89; p = 0.07). Conversion to thoracotomy was comparable (MD, 0.69; 95% CI, 0.37 to 1.28; p = 0.24). Hospital stay did not differ significantly (MD, 0.13; 95% CI, -0.58 to 0.84; p = 0.74). Operation time remained similar (MD, -4.63; 95% CI, -10.93 to 1.67; p = 0.15). The VATS procedure enabled significantly greater lymph node dissection (MD, 1.83; 95% CI, 0.16 to 3.49; p = 0.03). Chest drainage duration showed no difference (MD, 0.00; 95% CI, -0.27 to 0.27; p = 1.00).

CONCLUSION: In NSCLC, RATS and VATS yield comparable outcomes, with RATS potentially reducing blood loss and VATS allowing more lymph node dissection. Further high-quality RCTs are needed to clarify the optimal approach for NSCLC.

PMID:41369812 | DOI:10.1245/s10434-025-18862-w