To assess a deep learning (DL) model using portal-venous phase CT for discriminating colorectal cancer liver metastasis (CRLMs) and hemangiomas (HMs). Colorectal cancer (CRC) patients diagnosed with CRLMs or HMs at two medical centers from January 2018 and April 2024 were retrospectively included. Lesions were automatically segmented using TotalSegmentator. DL models, DenseNet-201 and ResNet-152, were trained to classify CRLMs and HMs. Their performance, measured by AUC, was evaluated on validation and test sets. Subgroup analyses were conducted for lesions ≤ 10 mm (subcentimeter) and 10–30 mm. Radiologists’ diagnostic performance with and without DL assistance was compared using a multi-reader multi-case analysis. 534 CRLMs (134 CRC-patients; median, 60 years) and 262 HMs (154 CRC-patients; median, 62 years) were divided into the training, validation and test set. The Dice coefficients of TotalSegmentor for automatically segmenting subcentimeter and 10–30 mm lesions were 0.692 ± 0.099 and 0.861 ± 0.033, respectively (p < 0.01). ResNet-152 model achieved AUCs of 0.875 (95

Background:Early prediction of efficacy of bevacizumab-combined chemotherapy in colorectal liver metastases (CRLM) remains challenging. This multicenter study aimed to develop and validate a multivariable computed tomography (CT)-based Delta-radiomics model to enable early and accurate prediction of treatment efficacy. Methods:We retrospectively analyzed consecutive patients with CRLM treated with bevacizumab-combined chemotherapy at three institutions from January 2018 to January 2023. According to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1, the therapeutic response of liver metastases and patient efficacy after 6 months of treatment were evaluated. The initial texture features were extracted from baseline and 2-month CT images to calculate temporal texture features (Ratio, Delta, DeltaABS). Eight logistic regression models using clinical and texture features were developed to predict the 6-month therapeutic response of liver metastases. Model performance was evaluated using area under the curve (AUC), calibration curves and decision curve analyses. Overall survival (OS) was analyzed using Kaplan-Meier curves and Cox regression. Results:A total of 90 patients and 255 liver metastases were included, with 133 liver metastases (52.16%) classified as responsive and 52 patients (57.78%) classified as responders. The Ratio, Delta and COMB models demonstrated superior performance in predicting the therapeutic response of liver metastases, with AUC ranging from 0.858 to 0.956 (training), 0.891 to 0.899 (internal validation), and 0.833 to 0.922 (external validation) across these models. The calibration and decision curves demonstrated that the prediction probabilities of the three models were highly consistent with the observed results and had good clinical utility. Cox regression analysis identified patient efficacy as the sole independent predictor of OS (P=0.002). Conclusions:The multivariable CT-based Delta-radiomics model demonstrates excellent performance in the early prediction of treatment efficacy of bevacizumab-combined chemotherapy in patients with CRLM, providing a novel tool for guiding personalized treatment strategies and early therapeutic assessment.

BACKGROUND:Intracranial solitary fibrous tumors (ISFTs) are frequently misdiagnosed as meningiomas on MRI. Existing MRI signs have been evaluated individually, without a standardized and externally validated diagnostic framework. PURPOSE:To develop and validate an MRI-based diagnostic score for intracranial solitary fibrous tumors (ISFT-DS) for the preoperative identification of ISFTs. STUDY TYPE:Retrospective multicenter diagnostic accuracy study. POPULATION:Five hundred with ISFTs (mean age, 46.4 ± 11.7 years; 269 men), 250 with low-grade meningiomas (mean age, 57.0 ± 11.1 years; 51 men), and 250 with high-grade meningiomas (mean age, 49.6 ± 14.3 years; 108 men). The development and external validation cohorts included 840 (84%) and 160 (16%) patients, respectively. FIELDSTRENGTH/SEQUENCE:Spin-echo T1-weighted imaging (T1WI), fast or turbo spin-echo T2-weighted imaging, and contrast-enhanced spin-echo T1-weighted imaging. ASSESSMENT:Three potential MRI features-T1WI hyperintensity, honeycomb-like cystic change, and trans-compartmental growth pattern-were assessed by three neuroradiologists and combined into the 0-3-point ISFT-DS. Diagnostic performance and the effects of standardized training in six readers were evaluated. STATISTICAL TESTS:Sensitivity, specificity, accuracy, kappa statistics, receiver operating characteristic analysis, DeLong tests, χ2and Fisher exact tests, analysis of variance, McNemar test, Kruskal-Wallis tests. p < 0.05 indicated statistical significance. RESULTS:At an ISFT-DS cutoff of ≥ 1, sensitivity/specificity were 81.4%/83.3% in the development cohort and 76.3%/86.3% in the external validation cohort, respectively. Specificity reached 98.3%/98.8% at ≥ 2 and 100%/100% at 3. Interreader agreement was good to excellent for the three MRI features and the ISFT-DS (κ = 0.80-0.87). Standardized reader training on the ISFT-DS increased the overall AUC from 0.67 to 0.80 (ΔAUC = 0.13), with larger gains for Readers 1 and 2 (ΔAUC = 0.21 and 0.15, respectively). DATA CONCLUSION:The ISFT-DS provides a reproducible and teachable MRI-based scoring system for standardized preoperative identification of ISFTs. EVIDENCE LEVEL:3. TECHNICAL EFFICACY:Stage 2.

BackgroundPlastic bronchitis (PB) is a complication of Mycoplasma pneumoniae pneumonia (MPP) in children, characterized by bronchial cast formation and airway obstruction. Recurrence after bronchoscopic cast removal can occur, and tools for early risk prediction are limiting. This study aimed to develop and externally validate a nomogram for predicting recurrent PB in children with MPP after initial bronchoscopic cast removal.MethodsThis retrospective multicenter cohort study included pediatric patients with MPP complicated by PB. Patients from one center were randomly divided into a training cohort and an internal validation cohort at a ratio of 8:2, while an independent cohort from another tertiary hospital served as the external validation chort. Candidate predictors were screened using least absolute shrinkage and selection operator (LASSO) regression, followed by multivariable logistic regression to identify independent predictors and develop the prediction model. Model performance was evaluated by discrimination, calibration, and clinical utility using receiver operating characteristic (ROC) curves, calibration plots, the Hosmer-Lemeshow goodness-of-fit test, and decision curve analysis (DCA).ResultsA total of 352 children with MPP complicated by PB were included, with 281 patients in the training cohort and 71 in the internal validation cohort. An additional 44 patients from an independent center formed the external validation cohort. Four variables were identified as independent predictors of PB recurrence: serum albumin (ALB), fever persisting for ≥72 h after the first bronchoalveolar lavage (BAL), atelectasis, and bronchial casts involving ≥2 lung lobes. These predictors were incorporated into a nomogram. The model showed discrimination across the three cohorts, with an area under curve (AUC) of 0.853 in the training cohort, 0.833 in the internal validation cohort, and 0.811 in the external validation cohort. Calibration plots showed agreement between predicted and observed risks, and decision curve analysis indicated a higher net benefit than the treat-all and treat-none strategies across a range of threshold probabilities.ConclusionWe developed and externally validated a nomogram to predict PB recurrence in children with MPP using routinely available clinical and bronchoscopic variables. The model showed consistent discrimination and calibration. Its potential clinical utility for risk stratification and post-procedural monitoring requires further prospective evaluation.

Abstract HEC-922 is a novel bispecific agonistic antibody targeting Cadherin-17(CDH17) and the immune agonistic receptor 4-1BB for the treatment of CDH17-positive tumors, particularly gastrointestinal tumors. CDH17 is a tumor associated antigen highly expressed in various tumors, including colorectal cancer, gastric cancer, and neuroendocrine tumors. Activating monoclonal antibodies against 4-1BB have shown clinical efficacy but was limited by systemic toxicity. The design of HEC-922 enables CDH17-dependent agonism of 4-1BB in the presence of CDH17 positive tumor cells, enabling tumor specific T cell activation while minimizing systemic immune toxicity. A high affinity nanobody against CDH17 and a nanobody against 4-1BB were identified for the construction of HEC- 922 containing an ADCC-silenced Fc. HEC-922 shows potent co-binding to both CDH17 and 4-1BB targets and is cross-reactive to the Rhesus macaque. In an assay using 4-1BB high expression 293 cells with NFkB-luciferase reporter, HEC-922 mediated 4-1BB activation in the presence of CDH17 positive cells but not CDH17 negative cells. HEC-922 demonstrated effective tumor growth inhibition in xenograft models of CDH17 positive tumor lines, restored the function of immune cells, reduced the proportion of exhausted T cells, and achieved a sustained and potent anti-tumor effect. Initial drug feasibility and toxicity studies results of HEC-922 were favorable. Overall result demonstrated that HEC- 922 is a promising candidate for CDH17 positive cancer immunotherapy.HEC-921, another bispecific antibody developed on the same 4-1BB platform with Ly6G6D as the targeted tumor antigen, has completed Dose Range Finding (DRF) study in cynomolgus monkeys, with a no-observed-adverse-effect level (NOAEL) of 100 mg/kg, validating the safety of the 4-1BB platform. Citation Format: Junji Dong, Shushan Lin, Zhou Linjun, Jiang Qiuyue, Chen Cangsha, He Shuiqing, Xiang Li, Ju Peng, Xiaohui Li, Cai Zhao, Ming Li, Xiaoping Li, Stewart Leung, . HEC-922:A CDH17-4-1BB bispecific antibody targeting CDH17 positive tumors shows potent anti-tumor activity [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2634.

Purpose To investigate whether the spectral computed tomography (CT) radiomics may predict the grade of esophageal varices (EV). Materials and Methods We retrospectively collected 301 cases with cirrhosis and EV and randomly divided them into training and test sets. Patients' clinical data, conventional enhanced CT characteristics, spectral CT 60 keV single-energy images, and iodine-based images were retrieved from the Picture Archiving and Communication Systems. Through comprehensive statistical analyses, including univariate analysis, correlation assessment, Least Absolute Shrinkage and Selection Operator regression, and multivariate regression analyses, the factors most related to EV grade were selected to construct the models. Predictive performance was evaluated by the receiver operating characteristic curve. A calibration curve was used to show the degree of fit between the nomogram and the actual results. The clinical utility of the model was evaluated using decision curve analysis (DCA). Results The diameters of the left gastric vein and EV were independent predictors of EV grade. Five out of 1,896 CT radiomics characteristics were correlated with EV grade. Six models were constructed. The model integrated with conventional enhanced CT and spectral CT radiomics characteristics performed best. The area under the curve in the training and test sets were 0.812 and 0.821, respectively. The calibration curve showed that this model had the highest agreement between observation and prognosis. The DCA found that this model provided the most clinical net benefit. Conclusion The comprehensive model based on enhanced CT and spectral CT radiomics performed best in predicting the grade of EV and may be used as a reference for clinical decision-making.

To compare the clinicopathological and magnetic resonance imaging (MRI) characteristics and prognostic implications between molecular glioblastoma (mol-GBM) and histological glioblastoma (hist-GBM) according to the 2021 WHO classification of central nervous system tumors (CNS5), we conducted a retrospective study of 105 patients with IDH-wildtype gliomas diagnosed between 2016 and 2022. Patients were reclassified into hist-GBM (n = 70) and mol-GBM (n = 35), with mol-GBM defined by TERT promoter mutation. Comparative analyses of clinical, molecular, and radiological features were performed. Overall survival (OS) was assessed in a subgroup of 96 patients using Kaplan-Meier and multivariable Cox regression models. Patients with mol-GBM were significantly younger (median age: 53 vs. 55.5 years, p = 0.046, median difference = - 2.5, 95% CI: -7.434 to 2.434), exhibited lower Ki-67 proliferation indices (30 vs. 40%, p = 0.002, median difference = - 10.0, 95% CI: -16.638 to - 3.362), and had lower p53 mutation rates (65.6 vs. 84.1%, p = 0.040, OR = 0.367, 95% CI: 0.138 to 0.973). MRI analysis demonstrated that mol-GBM was associated with less extensive peritumoral edema (major edema: 54.3 vs. 68.6%; p = 0.006) and lower tumor heterogeneity grades (Grade 3: 20.0% vs. 44.2%, p = 0.001, OR = 4.360, 95% CI: 1.950 to 9.746). No significant differences were observed in quantitative enhancement parameters or apparent diffusion coefficient (ADC) values. Survival analysis revealed a non-significant trend toward improved OS in the mol-GBM group (median OS: 17.0 vs. 13.0 months, log-rank p = 0.091). In multivariable analysis, GBM subtype was not an independent predictor of survival (hazard ratio [HR] = 0.829, 95% confidence interval [CI]: 0.432 to 1.591, p = 0.573). In conclusion, TERT promoter-mutant mol-GBM is characterized by a distinct clinicoradiological profile, including younger age at diagnosis and less aggressive imaging features. While these findings may aid preoperative suspicion, they do not confer independent prognostic value, underscoring the necessity of integrated molecular diagnosis to ensure appropriate treatment intensity and prevent undertreatment.

The value of a deep learning (DL) model in distinguishing intracranial solitary fibrous tumors (ISFTs) from angiomatous meningiomas (AMs) and predicting overall survival (OS) of patients with ISFTs have not been systematically assessed. The aim of this study was to develop and validate an MRI-based DL model for distinguishing ISFTs from AMs and predicting OS for patients with ISFTs. (Transformer + Clinic) and clinical models were developed and validated on retrospectively collected preoperative MRI scans of patients with ISFTs and AMs diagnosed between January 2008 and January 2023 at primary cohort (PC) and external validation cohort (EVC). We randomly selected 139 ISFT patients to form a follow-up cohort. The model with the highest mean area under curve (AUC) of receiver operating characteristic (ROC) on both cohorts was identified as optimal model (OM). The follow-up cohort were stratified into high- and low-risk groups based on a fixed cutoff calculated by the OM. The OM (Stepglm[both] + GBM) in (Transformer + Clinic) models outperformed the OM (Lasso + GBM) in clinical models in distinguishing ISFTs from AMs on EVC, with an AUC of 0.936 (95

OBJECTIVE:Coronary artery disease (CAD) progression is directly associated with major adverse cardiovascular events and death. This study aimed to construct a pericoronary adipose tissue (PCAT) radiomics model to predict subsequent progression in patients with CAD. METHODS:Data from 116 patients who had at least 2 coronary computed tomography angiography (CCTA) exams between March 1, 2020, and August 30, 2022, were collected at our institution. Obstructive stenosis, CAD-RADS classification, segment involvement score (SIS), and segment stenosis score (SSS) were noted. The radiomics features of the proximal to the left anterior descending artery, left circumflex artery, and right coronary artery were extracted on CCTA images using fully automated software. According to CAD-RADS, SIS, and SSS, non-progression was identified in 96, 80, and 72 patients and progression was identified in 20, 36, and 44 patients, respectively. All patients were randomly divided into the training and testing cohorts in a 7:3 ratio. Cox regression models were constructed based on PCAT radiomics signatures, and their predictive abilities were measured using receiver operating characteristic curves. RESULTS:We included 116 patients (age 58.00 [53.25, 64.00] years; 78 [67.20%] were male). After screening, 16 PCAT radiomics features were identified as being significantly related to CAD progression. The Cox regression models had area under the curve values of 0.841, 0.838, and 0.725 in the training cohort and 0.818, 0.817, and 0.851 in the testing cohort, respectively, to predict 2-year CAD-RADS, SIS, and SSS progression. CONCLUSIONS:PCAT-based radiomics models demonstrated promising performance in predicting subsequent CAD progression. ADVANCES IN KNOWLEDGE:PCAT-based radiomics signatures derived from coronary CT angiography provided incremental predictive value for CAD progression beyond conventional imaging markers (CAD-RADS, SIS, SSS), and may serve as noninvasive imaging biomarkers for individualized risk stratification.

To investigate the diagnostic value of various indicators and models for the differential diagnosis of non-muscle-invasive bladder urothelial carcinoma (NIBUC) and cystitis glandularis (CG). A retrospective analysis was performed using clinical and spectral computed tomography (CT) data from consecutive patients at Lanzhou University Second Hospital between January 2022 and January 2024. All patients underwent unenhanced arterial- and venous-phase spectral CT. Regions of interest were manually placed on the largest cross-section of the lesion. Iodine concentration (IC) and 40–70 keV monoenergetic CT values were measured in arterial and venous phases. The slope of the spectral CT-mono-energetic curve (40 and 70 keV; λHU) was calculated. Logistic regression analysis was used to construct two predictive models: one combining significant clinical indicators + conventional CT, and the other combining clinical indicators + conventional CT. +spectral CT parameters. The diagnostic performance was evaluated using receiver operating characteristic curve analysis. Thirty-nine patients with NIBUC and 21 patients with CG, all pathologically confirmed by surgery, were included. Differences in age, lower urinary tract symptoms, and maximum lesion diameter were significant between the two groups (all P < 0.05). The IC and the slope of the spectral CT monoenergetic curve in the arterial and venous phases were significantly higher in the NIBUC group than in the CG group (all P < 0.05). The arterial-phase spectral curve slope showed good diagnostic performance among the arterial- and venous-phase IC and spectral curve slopes (area under the curve [AUC] = 0.879). The combined model CI-CCT-λHU, incorporating statistically significant clinical indicators, conventional CT parameters, and arterial phase spectral curve slope, achieved the best diagnostic accuracy (AUC = 0.919). The arterial phase spectral curve slope provided good diagnostic efficacy for differentiating NIBUC from CG. A CI-CCT-λHU model combining clinical indicators, conventional CT, and arterial phase spectral curve slope can further improve the accuracy of differential diagnosis for these two lesions.

OBJECTIVE:To investigate the correlation between left atrial (LA) morphology and left ventricular (LV) systolic function in patients with atrial fibrillation (AF). METHODS:We studied 1 066 patients admitted to our institution between December 2018 and December 2022 for AF. All patients underwent cardiac computed tomography angiography (CTA) and transthoracic echocardiography. LV systolic function was assessed using transthoracic echocardiography. On cardiac CTA images, we measured the LA diameter (LAD) and quantified LA morphology using the fractal dimension (FD). The primary focus was association between the novel LA FD and LV systolic function, while LAD was included as a conventional comparator. RESULTS:Correlation analysis showed that LAD was negatively associated with the ejection fraction (EF) (r=-.175, P < .001) and fractional shortening (FS) (r=-.172, P < .001) and positively associated with end-systolic volume (ESV) (r = .090, P = .003). LA FD was negatively correlated with the EF (r=-.092, P = .003) and FS (r=-.083, P = .007) and positively correlated with end-diastolic volume (EDV) (r = .086, P = .005) and ESV (r = .094, P = .002). The restricted cubic spline (RCS) curves with four knots demonstrated a nonlinear correlation between LA FD and ESV (overall P = .046, nonlinear P = .035) and EDV (overall P = .002, nonlinear P < .001) and a linear relationship with EF (P = .037, P = .125). Specifically, EF demonstrated a continuous inverse linear association with increasing LA FD, whereas EDV and ESV followed a complex multiphasic pattern rather than a simple monotonic trend. CONCLUSIONS:The correlation between LV systolic function and LA FD suggests that LA shape may also be used as a tool to assess LV systolic function in patients with AF. ADVANCES IN KNOWLEDGE:In addition to LAD, LA FD may also provide clues to altered LV systolic function in this patient population.

While multimodal large language models (LLMs) demonstrate significant potential in healthcare applications, their clinical utility is difficult to appraise. Current evaluations of medical-assisting LLMs are often limited by sparse human expertise, narrow specialty scope, and reliance on multiple-choice benchmarks or synthetic vignettes, which can inflate performance and obscure clinical utility. We conducted a multicenter, multidisciplinary study in which more than 400 physicians—spanning seven specialties, varied experience levels, and multiple geographic settings—evaluated LLM-generated free-text responses to real, de-identified clinical cases. In a matched-control design, we also deployed an equivalent number of AI agents configured to mirror physician characteristics to examine whether automated evaluators can supplement or replace human assessment. Our results demonstrated that physician assessments exhibited substantial heterogeneity by clinical seniority and practice environment, leading to notable shifts in relative model rankings across cohorts. While AI agents delivered highly efficient, directionally aligned assessments, they did not fully capture the nuances of human clinical judgment and could not substitute for physician-centered evaluation. Instead, they promise assistive tools that can triage or pre-screen outputs to reduce human burden.

BACKGROUND:Predicting postoperative complications (POCs) in gastric cancer (GC) patients is increasingly important in clinical practice. Computed tomography (CT)-derived body composition (BC) parameters, reflecting nutritional and physiological status, have emerged as potential prognostic indicators. This systematic review and meta-analysis aimed to evaluate the association between preoperative muscle and fat parameters and POCs in GC patients. METHODS:A comprehensive search of PubMed, Web of Science, and Embase was conducted up to December 31, 2024. Two reviewers independently selected studies, extracted data, and assessed quality using the Quality in Prognosis Studies tool. This review specifically focused on studies that evaluated CT-derived BC parameters as independent prognostic factors through multivariable analyses. The primary outcome was the incidence of POCs within 30 days of radical gastrectomy. Pooled risk ratios (RRs) with 95% confidence intervals (CIs) were calculated using a random-effects model for parameters where quantitative synthesis was feasible. RESULTS:Out of 4736 identified records, 21 studies were included. All included studies provided adjusted effect estimates for BC parameters from multivariable analyses. Eight BC measures were assessed, including three muscle-related, two fat-related, and three composite metrics. Only sarcopenia defined by skeletal muscle index (SMI) met the criteria for meta-analysis. Pooled analysis of these adjusted estimates confirmed that sarcopenia was significantly and independently associated with POCs (RR = 2.73, 95% CI: 1.83-4.08, P = 0.010, I2 = 66%). CONCLUSIONS:CT-derived sarcopenia, defined by SMI, is an independent risk factor associated with increased POCs in GC patients. However, variability in measurement methods and outcome definitions limits the strength and clinical applicability of current evidence. Future studies should standardize BC assessment and reporting to better guide surgical risk stratification.

Adenoid cystic carcinoma (ACC) of the uterine cervix is a rare primary malignancy. A 60-year-old woman presented with a >1 year history of intermittent upper abdominal pain and discomfort, which had worsened over the preceding 4 days. Ten years earlier, the patient presented with irregular vaginal bleeding. Pelvic ultrasonography revealed a homogeneous myometrium, a clearly defined endometrium, and an indistinct echo pattern in the endocervical canal; no significant abnormalities were detected in either adnexa. Endometrial curettage and pathological biopsy confirmed ACC of the uterine cervix. The patient subsequently underwent total hysterectomy with bilateral oophorectomy, but declined postoperative adjuvant chemotherapy or radiotherapy. During the current admission, contrast-enhanced computed tomography of the entire abdomen revealed multiple, round, and slightly hypodense lesions in the right hepatic lobe, most of which were located near the liver margin and exhibited mild peripheral enhancement. Multiple soft tissue nodules of varying size were noted in the peritoneum and greater omentum, exhibiting heterogeneous and marked enhancement. Contrast-enhanced computed tomography of the head, neck, and chest revealed no significant abnormalities. Finally, fine-needle aspiration biopsy of the peritoneal nodule was performed, and metastasis of ACC of the uterine cervix was diagnosed. Palliative treatment was initiated; however, the patient died of multiple organ failure 1 month after treatment. This case indicates that ACC of the uterine cervix may develop distant metastasis as late as 10 years after initial diagnosis without adjuvant radiotherapy or chemotherapy. Clinicians should raise awareness of this disease, strengthen long-term follow-up, and perform timely imaging evaluations when symptoms occur.

Objective The 2021 WHO classification redesignates pituitary adenomas as pituitary neuroendocrine tumors (PitNETs), incorporating transcription factor profiling for subtype stratification. Given the current diagnostic challenges in distinguishing high- and low-risk PitNETs, we investigated whether MRI features combined with clinical biomarkers could improve preoperative risk stratification. Methods This multicenter retrospective study analyzed 548 histopathologically confirmed PitNET cases (training set: n = 319; test set: n = 138 from Center 1; validation set: n = 91 from Center 2). Comprehensive clinical, endocrinological, and MRI parameters were evaluated through logistic regression to construct a predictive model. Diagnostic performance was quantified using area under the ROC curve (AUC), supplemented by calibration plots and decision curve analysis (DCA) to assess clinical utility. Results Significant intergroup differences (all p < 0.05) were observed between high- and low-risk PitNETs: patient age, maximal tumor diameter (p < 0.01), growth hormone (GH), prolactin (PRL), insulin-like growth factor-1 (IGF-1) levels, tumor margin irregularity, optic chiasm compression (p < 0.001), circumferential carotid encasement, and cavernous sinus invasion. Multivariate analysis identified age (OR = 1.04, 95%CI 1.02–1.07), tumor diameter (OR = 1.15, 95%CI 1.08–1.22), PRL (OR = 1.01, 95%CI 1.00-1.02), and IGF-1 (OR = 1.003, 95%CI 1.001–1.005) as independent predictors. The integrated model achieved an AUC of 0.803 (95%CI 0.703–0.903) on external validation set, with excellent calibration and favorable decision curve net benefit. Conclusions Nomogram by integrating clinical and MRI features can be used as a reliable tool to predict risk status in patients with PitNETs. After further external validation, this will help neurosurgeons make critical decisions regarding surgical or alternative treatment strategies for PitNETs.

To evaluate the predictive value of whole-tumor iodine density (ID) histogram parameters and resection margin distance for early recurrence (ER) after curative resection of hepatocellular carcinoma (HCC). This retrospective study included patients with HCC who underwent R0 resection and received preoperative spectral CT scans. Patients were categorized into ER+ (n = 42) and ER− (n = 43) groups. Independent predictors of recurrence-free survival (RFS) were identified using multivariate Cox regression analysis. The performance of the prediction model was assessed using time-dependent receiver operating characteristic (td-ROC) curves, calibration and decision curves analysis. Kaplan-Meier analysis was used to evaluate differences in RFS between groups. Multivariate Cox regression identified Max, Skewness, microvascular invasion (MVI), and resection margin distance as independent risk factors for ER. Kaplan-Meier analysis revealed significantly shorter mean RFS in patients with MVI+ (12.34 months vs. 29.02 months), extremely narrow margin (9.54 months) and narrow margin (15.42 months) compared to wide margin (28.41 months), high Max (≥ 2041.00 vs. <2041.00; 15.15 vs. 26.13 months), and high Skewness (≥ 0.22 vs. <0.22; 16.83 vs. 23.62 months) (all P < 0.05). Whole-tumor ID histogram parameters (Max and Skewness) and clinicopathological factors (MVI and resection margin distance) are independent predictors of ER. These factors allow effective stratification of RFS and may guide individualized postoperative management. Whole-tumor iodine density histogram features and resection margin distance provide independent predictors of early recurrence after hepatectomy in HCC, enabling improved risk stratification and guiding individualized postoperative management.

OBJECTIVES:Accurately predicting meningioma brain invasion preoperatively helps to select the appropriate surgical approach and predict prognosis, but there are few imaging features that are sufficient for discriminating it alone. We investigate the joint MR imaging features and apparent diffusion coefficient (ADC) to predict the risk of brain invasion of meningiomas preoperatively. METHODS:In this retrospective study, 143 patients (invasion group:51, non-invasion group: 92) diagnosed with meningioma by histopathology were included. The maximum (ADCmax), minimum (ADCmin) and mean (ADCmean) values of ADC and the mean ADC values of a comparative ROI in the normal appearing white matter (ADCNAWM) were calculated. Differences between clinical features, MRI morphological features, and all ADC values were assessed by Pearson's chi-square test and Kruskal-Wallis rank-sum test. Stepwise logistic regression analysis was used to select the optimal features and construct a prediction model. Furthermore, A nomogram was used to predict the risk of brain invasion, and a decision curve was used to verify the clinical utility of the nomogram. RESULTS:According to stepwise logistic regression analysis, we found that sex, maximum diameter, peritumoral edema and ADCmin were closely related to brain invasion in meningioma. The model of the above four variables has the optimal discriminative ability to predict brain invasion, with an AUC of 0.924 (95 % CI, 0.879-0.969) and a sensitivity of 92.2 % (95 % CI, 74.5%-98.0 %). CONCLUSIONS:Combining clinical features, MRI morphological characteristics and ADCmin, the model exhibits excellent discriminatory ability and high sensitivity, which can be used for predicting the risk of brain invasion of meningiomas.

BACKGROUND:Pneumonia remains the predominant cause of childhood mortality and morbidity globally. While various imaging modalities have been employed for paediatric pneumonia diagnosis, the diagnostic accuracy remains inadequately characterised. OBJECTIVE:To systematically evaluate and compare the diagnostic accuracy of available imaging modalities for paediatric pneumonia through both diagnostic test accuracy (DTA) meta-analyses and network meta-analysis (NMA). METHODS:PubMed, Embase, Cochrane Library and Web of Science were searched up to March 2025. The risk of bias was graded using Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). Diagnostic accuracy measures were pooled using random-effects DTA meta-analyses, while relative diagnostic performance was compared through NMAs. RESULTS:81 studies published in 22 countries since 2008 were included, with a total of 34 625 children. In most of the studies, there was an unclear risk of bias. When clinical examination served as reference test, lung ultrasound demonstrated high diagnostic accuracy with sensitivity of 0.91 and specificity of 0.93. NMAs showed superior overall diagnostic performance of computer-aided chest radiography compared to lung ultrasound across all indexes except specificity, where there was no difference in sensitivity or specificity between the two. Meta-regression identified study design and pneumonia type as significant modifiers of diagnostic sensitivity. CONCLUSION:This comprehensive analysis provides robust evidence supporting the clinical utility of computer-aided chest radiography and lung ultrasound for paediatric pneumonia diagnosis. However, insufficient evidence precludes definitive conclusions regarding other computer-aided modalities. Future high-quality comparative studies are needed to validate these findings in diverse clinical settings and evaluate emerging imaging technologies.

Background:The prognosis of rectal cancer is closely related to its clinicopathologic features. Accurate preoperative assessment of these features is crucial for treatment planning and prognosis prediction. The apparent diffusion coefficient (ADC), derived from diffusion-weighted imaging (DWI), has shown potential as a noninvasive imaging biomarker for evaluating tumor characteristics. This study aimed to explore the relationship between ADC values and the clinicopathological features of rectal cancer. Methods:We retrospectively recruited 97 eligible patients with rectal adenocarcinoma who underwent magnetic resonance imaging (MRI) and surgical resection at our institution between January 2023 and December 2023. Each patient was evaluated for the presence of extramural vascular invasion (EMVI) or circumferential resection margin (CRM) on MRI, and the mean (ADCmean), minimum (ADCmin), and maximum (ADCmax) ADC values were calculated. Moreover, the relationship between the ADC values and clinicopathological features, including tumor stage, histologic grade, lymphovascular invasion, perineural invasion, and lymph node metastasis, were statistically analyzed. Results:Among 97 patients with rectal cancer, the mean age was 61.40±10.46 years and 60 (61.9%) were males. ADCmean, ADCmin, and ADCmax were significantly lower in patients with EMVI or CRM than in those without EMVI or CRM (P<0.05). Pathologic T1-2 staging exhibited higher ADCmean (0.79±0.26 vs. 0.61±0.22, P=0.001), ADCmin (0.71±0.26 vs. 0.55±0.22, P=0.002) and ADCmax (0.89±0.26 vs. 0.75±0.22, P=0.004) compared with T3-4 staging. Highly and moderately differentiated tumors had higher ADCmean, ADCmin, and ADCmax than less-differentiated tumors (P<0.05). Patients with lymphovascular invasion, perineural invasion, and lymph node metastasis showed significantly lower ADCmean, ADCmin, and ADCmax than those without these conditions (P<0.05). ADCmean, ADCmin and ADCmax were negatively correlated with EMVI (r=-0.334, -0.340, -0.302), CRM (r=-0.362, -0.414, -0.276), pathologic T-stage (r=-0.324, -0.313, -0.276), histologic grade (r=-0.353, -0.352, -0.289), lymphovascular invasion (r=-0.405, -0.384, -0.421), perineural invasion (r=-0.428, -0.407, -0.265), and lymph node metastasis (r=-0.347, -0.316, -0.268) in rectal cancer. Conclusions:ADC values were negatively associated with different clinicopathological features of rectal cancer, suggesting their potential role as noninvasive imaging markers for preoperative tumor assessment.