Search Results (696)

Background and Objectives: The double burden of malnutrition (DBM), characterized by the coexistence of malnutrition and overweight within the same household, has become a significant public health concern in low- and middle-income countries. Tanzania is undergoing a nutritional transition marked by persistent child malnutrition alongside increasing maternal overweight. This study examined socio-demographic, maternal, and child-level factors associated with DBM among children under five years in Tanzania. Methods: This cross-sectional study used data from the 2022 Tanzania Demographic and Health Survey, including a weighted sample of 5744 children under five and their mothers aged 15–49 years. DBM was defined as the presence of child malnutrition, measured using the Composite Index of Anthropometric Failure (CIAF), in households where the mother was overweight or obese. Bivariate chi-square tests and binary logistic regression analyses were conducted in STATA 17. Adjusted odds ratios (AORs) with 95% confidence intervals (CIs) were estimated to identify predictors of DBM. Results: DBM was more prevalent in rural areas. Significant predictors included birth order (AOR = 0.611, p = 0.030), child sex (AOR = 0.708, p = 0.011), perceived birth size (AOR = 0.270, p = 0.004), child age (AOR = 0.474, p < 0.001), maternal age (AOR = 0.599, p = 0.045), and maternal education (AOR = 0.604, p = 0.035). Higher maternal education reduced the likelihood of DBM, while firstborn male and small-sized children were at greater risk. Conclusions: DBM in Tanzania is influenced by both biological and socio-demographic factors. Integrated, multi-sectoral interventions targeting maternal education, prenatal care, and optimal maternal nutrition are essential to reduce DBM and achieve global malnutrition reduction targets. Full article

Background: Glioblastoma exhibits profound intratumoral heterogeneity, with anatomically distinct tumor zones characterized by divergent molecular programs that drive therapy resistance. Whether magnetic resonance imaging (MRI)-derived radiomic features can capture these regional transcriptomic differences remains unknown. We aimed to determine whether subcompartment-level radiomic features associate with transcriptomic pathway enrichment scores derived from biologically approximate tumor zones. Methods: We matched 28 patients (mean age 58.5 years; 13/28 MGMT methylated) across the IvyGAP RNA-seq atlas and the IVYGAP-RADIOMICS datasets. Single-sample GSEA (ssGSEA) pathway scores were computed for 24 gene sets. Radiomic features (3920 per subcompartment) were reduced to 597. Nested leave-one-patient-out cross-validation (LOPO-CV) with Elastic Net served as the primary predictive analysis; linear mixed-effects models (LMM) provided exploratory associational analysis. Analyses used a biologically motivated but spatially non-co-registered zone-to-subcompartment mapping; all reported associations are zone-approximate. Results: Twenty-one of 24 pathways showed no predictive signal (R²_cv ≤ 0). Inflammatory Response (R²_cv = 0.185, 95% CI [0.071, 0.355], p = 0.008) was the only pathway supported by both the nested CV (FDR = 0.096) and the exploratory LMM (FDR = 0.024, ΔR² = 0.214 beyond subcompartment effects) analyses; the LMM association was robust to clinical covariate adjustment (likelihood ratio test p = 0.004). Angiogenesis (R²_cv = 0.209, 95% CI [0.028, 0.353], p = 0.006) reached nested CV significance (FDR = 0.096) but was not corroborated by the LMM (FDR = 0.445); it is therefore reported as a tentative single-framework signal requiring independent validation. T2-derived texture features were selected in 100% of folds for both pathways. Conclusions: Inflammatory Response is the only pathway supported by both analytical frameworks; Angiogenesis is a tentative nested-CV-only signal pending independent validation. The absence of signal for 21 of 24 pathways should not be interpreted as evidence of biological inaccessibility: at N = 28 (vs. N ≈ 240 required by Riley criteria), severe underpowering, attenuation from the non-spatial zone-to-subcompartment mapping, and methodological constraints each independently suffice to suppress real associations. Five of the 24 gene sets (the IvyGAP zone modules) are non-independent from the outcome data and cannot be interpreted as discovery. All reported associations are zone-approximate and may partly reflect macro-compartment (between-subcompartment) effects; validation in larger cohorts with spatially precise co-registration is essential. Full article

Background: Treatment adherence and outcomes for drug-resistant tuberculosis (DR-TB) continue to be subpar in rural South Africa, where structural health system limitations, comorbid conditions, and diverse resistance patterns make clinical management more challenging. This study aimed to assess how demographic, clinical, and programmatic factors, including a Community Engagement–Clinical Governance (CE–CG) implementation period, affect DR-TB treatment outcomes using explanatory predictive modeling. Methods: A retrospective cohort study was conducted using routine program data from 694 DR-TB patients. A complete-case analysis was performed for multivariable modeling (n = 282). Logistic regression and decision tree models were used to examine the relationships between treatment success and selected predictors, including age, sex, treatment regimen, resistance phenotype, comorbidities, and the CE–CG implementation period. Model discrimination and performance were evaluated using receiver operating characteristic (ROC) curves, pseudo-R² statistics, likelihood ratio tests, and multicollinearity diagnostics. Results: The cohort had a mean age of 40.7 years, and 58.8% of patients were male. Overall treatment success was 59.9%. Severe resistance phenotypes were rare (1.7%) but clinically significant. Comparative analysis showed no notable demographic or outcome differences between included and excluded patients, indicating minimal selection bias. In adjusted models, treatment initiation during the CE–CG implementation period was significantly linked to lower odds of treatment success (adjusted odds ratio [aOR] = 0.443; 95% CI: 0.240–0.818; p = 0.009). Severe resistance phenotypes were strongly negatively associated with treatment success (aOR = 0.303; p = 0.056). Logistic regression models had limited discriminatory ability (AUC: 0.523–0.548), while the decision tree model showed modest improvement (AUC: 0.626). Overall, the model’s explanatory power was limited (pseudo-R² = 0.029), although no evidence of multicollinearity was found. Conclusions: Programmatic implementation periods and resistance severity were important factors associated with treatment outcomes in this rural DR-TB cohort. Although model discrimination was modest and explanatory power was limited, the findings provide useful insights into structural and programmatic vulnerabilities that affect treatment success in real-world settings. Strengthening clinical governance, improving routine program documentation, and incorporating more granular adherence, social, and governance indicators into routine data systems may improve both program evaluation and future predictive modeling. Full article

Background/Objectives: Chronic endometritis (CE) is an underrecognized inflammatory disorder associated with infertility, recurrent pregnancy loss, and implantation failure. CD138 immunohistochemistry is widely used to detect endometrial plasma cells; however, background epithelial staining and interobserver variability may limit diagnostic precision. MUM1 (IRF4), a nuclear transcription factor expressed in plasma cells, has emerged as a potential complementary marker. We aimed to systematically evaluate the diagnostic performance of MUM1 immunohistochemistry relative to CD138-based histopathologic reference frameworks for chronic endometritis using a Bayesian meta-analytic framework. Methods: We conducted a systematic review and diagnostic test accuracy meta-analysis in accordance with PRISMA 2020 and PRISMA-DTA guidelines. MEDLINE, Embase, Scopus, and CENTRAL were searched from inception to 28 February 2026. Studies assessing MUM1 immunohistochemistry against predefined histopathologic reference frameworks for CE were eligible. Risk of bias was evaluated using QUADAS-2. A bivariate Bayesian random-effects model was applied to jointly estimate pooled sensitivity, specificity, likelihood ratios, and diagnostic odds ratio (DOR). Results: Six studies (n = 1574 women) were included in the qualitative synthesis, and four provided sufficient 2 × 2 data for quantitative pooling. The pooled sensitivity of MUM1 was 0.876 (95% credible interval (CrI): 0.536–0.976), and the pooled specificity was 0.853 (95% CrI: 0.653–0.930). The pooled positive likelihood ratio was 5.83 (95% CrI: 2.18–12.04), and the negative likelihood ratio was 0.15 (95% CrI: 0.03–0.56), corresponding to a DOR of 40.27 (95% CrI: 5.04–254.59). Credible intervals were wide, reflecting statistical uncertainty related to the limited number of studies and heterogeneity in diagnostic thresholds. Conclusions: MUM1 (IRF4) immunohistochemistry demonstrates potentially favorable comparative diagnostic performance relative to CD138-based reference frameworks, although substantial uncertainty remains due to limited and heterogeneous evidence. As an adjunctive nuclear marker, MUM1 may support histopathologic assessment of chronic endometritis; however, prospective head-to-head studies using harmonized diagnostic criteria and predefined plasma cell thresholds are required before routine implementation can be firmly recommended. Full article

Background/Objectives: Atrial fibrillation is the most common sustained arrhythmia and frequently occurs in the perioperative setting. However, its clinical relevance in neurosurgical patients remains poorly defined despite increased vulnerability related to brain injury, inflammation, and perioperative stress. This study aimed to determine whether newly detected postoperative atrial fibrillation (POAF) identifies a higher-risk profile and is associated with postoperative complications, resource utilization, and short-term mortality compared with patients remaining in sinus rhythm (SR). Methods: We conducted a single-center retrospective cohort study (October 2020–April 2025). Among 2619 neurosurgical procedures, only patients with both pre- and in-hospital post-procedure ECGs and no pre-existing arrhythmia were included. POAF was defined as atrial fibrillation detected on any post-procedure ECG. Outcomes were compared using Welch’s t-test, χ²/Fisher’s exact tests, and odds ratios (OR) with 95% confidence intervals (CI). This study was not designed to estimate the incidence of POAF but rather to compare outcomes within a selected ECG-screened subgroup. Results: A total of 171 patients met the inclusion criteria: 79 (46.2%) developed POAF and 92 (53.8%) remained in SR. Patients with POAF were older and had a higher burden of cardiometabolic comorbidities and were more likely to undergo craniotomy/trepanation and emergency procedures. Compared with SR, POAF was associated with higher rates of postoperative complications, longer ICU and hospital stay, lower likelihood of discharge home, and higher short-term mortality. These findings reflect a selected ECG-screened cohort and should not be interpreted as the incidence of POAF in the overall neurosurgical population. Conclusions: Newly detected POAF is associated with a higher-risk postoperative profile in neurosurgical patients. It clusters with greater comorbidity burden, more invasive and urgent procedures, increased complications, prolonged hospitalization, reduced likelihood of discharge home, and higher short-term mortality. These findings support further evaluation of rhythm surveillance and perioperative management strategies in higher-risk neurosurgical populations. Full article

This article provides a selective review of offline change-point detection methods, with a particular emphasis on weak change-points. The study of such weak changes plays a central role in establishing the asymptotic properties of test statistics designed for detecting fixed (non-vanishing) changes. In addition, it is crucial for analyzing the asymptotic behavior of change-point location estimators. We review the current state of the literature, identify key limitations, and outline promising directions for future research in this challenging setting. Full article

The oral cavity performs several functions that are essential for sustaining life and carrying out daily activities. Given the importance of maintaining functional abilities for healthy aging, exploration of the adverse health outcomes that may result from impaired oral function is needed. This cross-sectional study examined oral functional status in 176 community-dwelling older adults with preserved cognitive function and activities of daily living, and analyzed the association between oral hypofunction and physical frailty. Oral hypofunction was diagnosed using seven parameters. Frailty was assessed using the Korean version of the Fried Frailty Phenotype, which focuses on physical characteristics. Independent t-tests, one-way analysis of variance, the Kruskal–Wallis test, and multiple logistic regression analysis were conducted to examine the association between oral hypofunction and frailty. The prevalences of oral hypofunction and physical frailty were approximately 48% (n = 85) and 40% (n = 71), respectively. After adjusting for frailty-related factors, including sociodemographic and health-related characteristics, higher oral hypofunction scores were associated with an increased likelihood of physical frailty (odds ratio, 1.382; 95% confidence interval, 1.017–1.876). These findings suggest that maintaining the various functions of the oral cavity, including mastication, within normal ranges and restoring impaired oral abilities as early as possible may be relevant to healthy aging. Full article

Background: Human cysticercosis, caused by the larval stage (cysticerci) of the pork tapeworm Taenia solium, is an important zoonotic disease. The disease is prevalent in developing countries where porcine cysticercosis is common and undercooked pork is habitually consumed. Objective: This study aimed to develop an immunochromatography-based test kit for the rapid diagnosis of human cysticercosis using low-molecular-weight antigens purified from cyst fluid of the T. solium Asian genotype to detect specific IgG antibodies in whole blood. The kit was designated as “the cysticercosis whole-blood test kit (iCys WB kit).” Methods: It was evaluated under laboratory conditions using 164 whole-blood samples, of which 21 were from confirmed cysticercosis cases. The results of the iCys WB kit, which detects anti-T. solium (cysticercus) antibodies in simulated whole blood samples, were compared with results from corresponding human serum samples. Results: When using both sample types, iCys WB kit demonstrated an accuracy of 98.8%, a sensitivity of 91.7%, a specificity of 100%, a positive likelihood ratio of 0, a negative likelihood ratio of 0.083, and an ROC area of 0.96. The agreement between results obtained from simulated whole-blood and serum samples showed perfect concordance. Conclusions: The iCys WB kit is a valuable easy-to-handle diagnostic tool and may be applicable for supporting clinical diagnosis at the point of care. Full article

In clinical research, unilateral data and bilateral data are commonly collected when paired organs or body parts of people receive treatment. Existing models are often inadequate for the research of combined unilateral and bilateral data. Considering population heterogeneity, this paper proposes three statistical tests and sample size estimation methods for the relative risk ratio in stratified unilateral and bilateral data under Dallal’s model. We derive test statistics (i.e., likelihood ratio, Wald-type, and score statistics) and evaluate their performance in terms of type I error rates and powers. Then, sample size determination is performed using an iterative algorithm. Monte Carlo simulations demonstrate that the score test performs well across various parameter configurations. Moreover, the estimated powers for determining sample size based on the score test are closer to the actual empirical powers. Two real examples of otolaryngology and myopathy are provided to illustrate the effectiveness of the proposed methods. Full article

Background/Objectives: Healthcare-associated ventriculitis and meningitis (HAVM) is a life-threatening complication of neurosurgical procedures. Conventional cerebrospinal fluid (CSF) indices cannot reliably distinguish bacterial infection from sterile postoperative inflammation, and cultures are frequently delayed or negative. We conducted the first systematic review and meta-analysis to determine the pooled diagnostic accuracy of CSF heparin-binding protein (HBP) for HAVM and to establish clinically actionable decision thresholds. Methods: PubMed, Embase, the Cochrane Library, and China National Knowledge Infrastructure were searched from inception to 15 February 2026. Risk of bias was assessed using QUADAS-3. Sensitivity and specificity were pooled with a bivariate random-effects model, and heterogeneity was explored through subgroup analyses and metaregression. Thresholds were derived using likelihood ratio (LR)-based and diagmeta cutoff modeling. Results: Twelve studies (n = 1761) were included. Pooled sensitivity was 0.861 (95% confidence interval [CI]: 0.777–0.917) and specificity was 0.848 (95% CI: 0.781–0.897), with a positive LR (LR+) of 5.65 and a negative LR (LR−) of 0.164. At a 50% pretest probability, post-test probability was increased to 85% by a positive result and reduced to 14% by a negative result. Intracerebral hemorrhage cohorts showed lower accuracy (sensitivity: 0.675, specificity: 0.755), whereas brain tumor-predominant cohorts demonstrated the highest performance (sensitivity: 0.935, specificity: 0.922; p = 0.017). Thresholds of ≥41.3 (rule-in; LR+ ≥10) and ≤30.1 ng/mL (rule-out; LR− ≤0.1) defined clinically actionable decision zones. Conclusions: CSF HBP provides quantitatively defined rule-in and rule-out thresholds that meaningfully shift the post-test probability and may support antimicrobial decision-making in suspected HAVM. Prospective multicenter validation is warranted. Full article

Background/Objectives: Fibromyalgia is increasingly viewed as a disorder of central sensitization, involving altered nociceptive processing and dysregulated stress and affective neural systems. Evidence supports shared neurobiological mechanisms linking chronic pain, emotional distress, and affect regulation, including corticolimbic and hypothalamic–pituitary–adrenal axis alterations. However, predictive models evaluating psychological distress as markers of these brain-based processes remain scarce. This study aimed to internally validate a preliminary model of fibromyalgia diagnosis using self-reported distress indicators as proxies of central dysregulation. Methods: A case-control design study with 180 participants was performed. Medically diagnosed fibromyalgia cases were recruited via a pain facility or referrals, alongside geographically matched controls from the general population. Psychological variables were conceptualized as neurobehavioral indicators reflecting central sensitization and stress-system dysregulation. Predictors were selected using LASSO penalized regression with 10-fold cross-validation. Retained variables were re-estimated using logistic regression. Model performance was evaluated through Nagelkerke’s pseudo-$R^2$, a likelihood ratio test, and area under the curve (AUC). Internal validation was conducted via 1000-bootstrap resampling with calibration-slope-based shrinkage. Results: The final model included global psychological distress, positive affect, sex, and age ($R^2=0.359$, with good discrimination [AUC = 0.81; optimism-corrected AUC ≈ 0.79]). Higher distress and age were associated with increased odds of fibromyalgia. Conclusions: Self-reported psychological distress, particularly global distress and reduced positive affect, combined with sex and age, showed internal validity in predicting fibromyalgia diagnosis. These findings support the hypothesis that behavioral markers of emotional dysregulation may reflect underlying central sensitization and stress-system alterations implicated in chronic pain. Future research integrating psychological measures with neuroimaging and neuroendocrine markers may further clarify the neural mechanisms linking affective dysregulation and chronic pain vulnerability. Full article

Process monitoring methods play a crucial role in identifying equipment malfunctions and instrument failures, as well as in maintaining process safety and product quality. Selecting the right approach for fault detection and diagnosis is therefore vital. Several localization methods based on Kernel Principal Component Analysis (KPCA) exist, such as the partial localization approach, which is effective at detecting anomalies but does not always pinpoint faults precisely. This method often identifies a suspicious area or group of variables without isolating the exact source of the fault. In complex systems such as chemical reactors, it can produce false positives or incorrect localizations if the data are noisy or if the fault affects multiple correlated variables. Conversely, the reconstruction-based contribution approach, when integrated with Kernel Principal Component Analysis (KPCA), is both widely documented in the literature and highly effective for fault localization. This method first identifies anomalies using the Hotelling’s T² statistic and Q (squared prediction error) statistic, then analyzes the contributions of individual variables to these indices in order to isolate the fault. However, the convergence of the optimization algorithm using the T² index is not guaranteed. To address this limitation, we introduce RBC-KGLRT, a novel localization framework that integrates reconstruction-based contribution with KPCA and the Generalized Likelihood Ratio Test in its kernel form to improve both precision and reliability in localization tasks. This work transforms traditional KPCA and reduced-rank KPCA fault detection approaches—enhanced by the KGLRT metric—into a powerful fault localization solution through the reconstruction-based contribution (RBC) method. Its effectiveness is rigorously evaluated using the Tennessee Eastman Process (TEP), a widely recognized simulation benchmark in process control and chemical engineering. Full article

Finned tubes contribute to the heat transfer performance of heat exchangers by increasing the surface area; they also modify patterns within the flow around the tubes and thus increase the likelihood of flow-induced vibrations (FIVs), which can undermine structural integrity. The tradeoff between improved heat transfer and minimized vibration risks is thus of concern in the optimization of finned tube designs. This paper examines the vibration behavior of circular finned tubes fitted in a parallel triangular configuration when subjected to crossflow conditions with particular reference to the structural response as opposed to thermal performance. In this study, two tube bundles arranged in a 30° parallel triangular layout were tested. The test tube has pitch-to-diameter (P/D) ratios of 1.16 and 1.37 and fin densities of 3, 6, and 9. In this study, experiments were conducted in a low-speed closed-loop water tunnel, which also involved the fabrication of circular finned tubes, the preparation of test bundles, and vibration response measurements. The key parameters analyzed in this experiment were the vibration amplitude, damping, pitch ratio, and fin density. Based on the free-stream velocity range of 0.13–0.28 m/s in a 300 mm × 300 mm closed-circuit water tunnel (hydraulic diameter $D_h=0.3$ m), the Reynolds number ranged from 3.9 × 10⁴ to 8.4 × 10⁴ (water at 20 °C). The results of this experiment demonstrate that by increasing the fin density, the vibration amplitudes can be reduced, which also raises the critical velocities. Reducing the pitch ratio from 1.37 to 1.16 produced an onset of instability approximately 53% earlier than the onset of instability at the ratio of 1.37. The bandwidth of the pitch ratio of 1.16 at the same fin density of 9 was almost 45% lower than that at 1.37, which confirms that the system at 1.16 is much more unstable. In general, the 1.37 pitch ratio offers 3 times higher stability margins than those of 1.16 for the fin densities under study. The development of optimal finned tube heat exchanger designs that reduce flow-induced vibrations without sacrificing thermal performance is aided by these findings, which provide information on the relationship between the fin density, pitch ratio and vibration behavior. Full article

Background/Objectives: To compare the maternal, fetal, and neonatal outcomes of pregnancies undergoing amniocentesis with those undergoing non-invasive prenatal testing (NIPT), within a cohort of women with comparable clinical indications, aiming to evaluate differences in adverse outcomes in a risk-indicated population. Methods: In this retrospective cohort study, pregnancy outcomes of 2044 pregnant women who underwent amniocentesis and 7668 pregnant women who underwent NIPT were evaluated using single-center data. The analysis was restricted to pregnancies with normal/reportable test results and without structural or genetic anomalies. Pregnancy loss outcomes were evaluated in the full cohort, while perinatal outcomes were analyzed among cases with available delivery data (377 amniocentesis and 2063 NIPT cases). Pregnancy and perinatal outcomes, including miscarriage, intrauterine fetal demise (IUD), preterm birth (PTB), pregnancy-induced hypertensive diseases (PIHDs), gestational diabetes mellitus (GDM), intrahepatic cholestasis of pregnancy (ICP), low birth weight (LBW), small for gestational age (SGA), and low APGAR scores (<7), were evaluated. Multivariate logistic regression analysis was performed to adjust for potential confounding factors, and adjusted odds ratios (aORs) with 95% confidence intervals (CIs) were reported. Results: Amniocentesis was associated with a significantly higher risk of an adverse outcome compared to NIPT in this risk-indicated cohort. The likelihood of miscarriage was significantly higher in the amniocentesis group (aOR: 1.91, 95% CI: 1.17–3.14, p = 0.025), as was the risk of IUD (aOR: 4.10, 95% CI: 2.05–8.20, p < 0.001). PTB risk was also increased (aOR: 1.96, 95% CI: 1.53–2.51, p < 0.001). LBW was significantly more prevalent in the amniocentesis group (aOR: 7.73, 95% CI: 5.40–11.05, p < 0.001), and the likelihood of delivering a SGA neonate was also increased (aOR: 1.45, 95% CI: 1.02–2.06, p = 0.040). A 1st-minute APGAR score < 7 was also more frequent in the amniocentesis group (aOR: 1.51, 95% CI: 1.06–2.16, p = 0.022), although the association with 5th-minute APGAR scores < 7 did not reach statistical significance (aOR: 1.45, 95% CI: 0.83–2.52, p = 0.193). Overall, the risk of composite maternal and perinatal adverse outcomes (aOR: 1.77, 95% CI: 1.41–2.22, p < 0.001) as well as composite fetal and neonatal adverse outcomes (aOR: 1.97, 95% CI: 1.50–2.58, p < 0.001) was significantly higher in the amniocentesis group compared to the NIPT group. No significant association was observed for PIHD, GDM, or ICP. Conclusions: Our findings showed that, apart from fetal loss, amniocentesis may be associated with adverse perinatal outcomes such as PTB, LBW, SGA and low APGAR scores. Full article

Background: Intrahepatic cholangiocarcinoma (ICC) is an aggressive primary liver malignancy with limited survival, largely due to delayed diagnosis, recurrence and limited effective therapeutic options. Radiomics- and artificial intelligence (AI)-based imaging models have emerged as promising tools to improve noninvasive detection and differentiation of ICC. We conducted a systematic review and meta-analysis to evaluate the diagnostic performance of radiomics-based AI models for ICC. Methods: A systematic search of PubMed, Embase, Scopus, and the Cochrane Library was performed from inception through 2025 in accordance with PRISMA guidelines. Studies assessing radiomics- or AI-based models derived from CT, MRI, PET, or ultrasound for differentiation of ICC from other hepatic lesions were included. Pooled sensitivity, specificity, positive likelihood ratio (PLR), and negative likelihood ratio (NLR) were estimated using a bivariate random-effects model. Study quality and risk of bias were assessed using the Radiomics Quality Score (RQS) and QUADAS-2 tools. Results: Twenty retrospective studies comprising 8746 participants were included. Across pooled validation and test datasets, radiomics-based AI models demonstrated a pooled sensitivity of 0.77 (95% CI, 0.69–0.84) and specificity of 0.88 (95% CI, 0.78–0.94) for differentiating ICC from non-ICC hepatic lesions. The pooled PLR was 6.81 (95% CI, 3.51–13.2), and the pooled NLR was 0.23 (95% CI, 0.09–0.61). CT-based models showed higher diagnostic performance compared with MRI and ultrasound. Subgroup and meta-regression analyses identified imaging modality, contrast phase, segmentation strategy, and validation approach as contributors to interstudy heterogeneity. The overall methodological quality demonstrated a mean Radiomics Quality Score (RQS) of 14.0 (range 11–24), corresponding to approximately 39% of the maximum achievable score. External validation cohorts were incorporated in 60% of the studies, although adherence to standardized feature reproducibility frameworks varied. Conclusions: Radiomics-based AI models demonstrate clinically meaningful diagnostic accuracy for noninvasive differentiation of ICC and may complement conventional imaging in preoperative evaluation. Prospective, multicenter studies with standardized imaging protocols and rigorous external validation are required before routine clinical adoption. Full article