OBJECTIVES:To develop and test a convolutional neural network model for automated segmentation of complicated cystic renal masses (cCRMs) on MRI. METHODS:This multicenter retrospective study analysed 210 cCRMs between October 2019 and May 2021, divided into training/internal validation (n = 150, Institution 1) and test sets (n = 60, Institutions 2-4). Comparative 3D V-Net and U-Net models were developed across 7 MRI sequences (T2-weighted, diffusion-weighted, apparent diffusion coefficient maps, unenhanced T1-weighted, and enhanced corticomedullary, nephrographic, and excretory phases images). A total of 14 models were developed, and 7 pairwise comparisons were performed between the 3D V-Net and U-Net models. Segmentation performance was evaluated using Dice similarity coefficient (DSC) and Hausdorff distance (HD), with subgroup analysis of small cCRMs (≤40 mm). RESULTS:In the test set, the excretory-phase V-Net (EPV-Net model) showed the highest DSC, and perform better than the corresponding U-Net (EPU-Net model) across all cCRMs (DSC: 0.74 ± 0.05 vs 0.70 ± 0.06, P < .001; HD: 27.41 ± 7.44 mm vs 39.18 ± 11.07 mm, P < .001) and the 35 small cCRMs subgroup (DSC: 0.74 ± 0.05 vs 0.70 ± 0.06, P < .001; HD: 27.48 mm ± 6.32 vs 38.72 ± 10.69 mm, P < .001). CONCLUSIONS:The 3D EPV-Net model demonstrated good segmentation accuracy, even for small lesions, supporting its clinical utility for cCRMs evaluation. ADVANCES IN KNOWLEDGE:This automated approach may streamline workflow compared to manual segmentation in cCRMs assessment.

Purpose To develop and evaluate a preoperative MRI-based model for predicting inferior vena cava (IVC) wall invasion in renal cell carcinoma (RCC) with IVC tumor thrombus (IVCTT) and to compare its performance with that of individual MRI features and radiologists' subjective assessments. Materials and Methods This single-center study with retrospective and prospective components included individuals who underwent or were scheduled to undergo surgery for RCC with IVCTT (retrospective training set, n = 173, January 2005-December 2023; prospective temporal validation set, n = 44, January 2024-September 2025). Histopathology served as the reference standard. Quantitative (tumor, vessel, and thrombus measurements) and qualitative (signal and morphologic characteristics) MRI features were assessed. Two fellowship-trained abdominal radiologists independently provided subjective assessments of IVC wall invasion, and interobserver agreement was assessed. Variables significant at univariable analysis were entered into multivariable logistic regression to identify predictors of IVC wall invasion. Diagnostic performance was compared using receiver operating characteristic (ROC) curve analysis and DeLong tests. Results A total of 217 individuals were included (mean age, 57 years ± 12 [SD], 166 male). Four independent predictors of IVC wall invasion were identified: bland thrombus (odds ratio [OR] = 3.32 [95% CI: 1.38, 8.03]), lumbar vein diameter (>5.25 mm) (OR = 2.64 [95% CI: 1.23, 5.69]), ipsilateral renal vein ostium diameter (>19.20 mm) (OR = 3.64 [95% CI: 1.73, 7.63]), and thrombus craniocaudal length (>46.95 mm) (OR = 3.08 [95% CI: 1.43, 6.63]). The multivariable model incorporating these predictors achieved area under the ROC curve (AUC) values of 0.81 (95% CI: 0.75, 0.88) and 0.84 (95% CI: 0.73, 0.96) in the training and validation sets, respectively, significantly outperforming the best individual MRI predictor (AUC = 0.71) and radiologists' subjective assessments (AUC = 0.66) (all P < .05). Conclusion The multiparametric MRI-based model demonstrated good discriminatory performance for predicting IVC wall invasion and outperformed individual MRI features and subjective radiologist assessment. Keywords: MR Imaging, Urinary, Kidney, Renal Cell Carcinoma, Magnetic Resonance Imaging, Inferior Vena Cava Tumor Thrombus, Venous Wall Invasion Supplemental material is available for this article. © RSNA, 2026.

To evaluate intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI) for stratifying early response to neoadjuvant immune checkpoint inhibitor and tyrosine kinase inhibitor (ICI-TKI) therapy in renal cell carcinoma (RCC) and explore correlations between IVIM‑DWI parameters and tumor immune cell infiltration. This retrospective exploratory analysis utilized prospectively collected data from a single‑center study enrolling patients with advanced, metastatic or unresectable RCC scheduled for neoadjuvant ICI‑TKI therapy. Two radiologists independently segmented primary tumors on baseline IVIM‑DWI to obtain the true diffusion coefficient (D), pseudo‑diffusion coefficient (D*), perfusion fraction (f), and standard apparent diffusion coefficient (ADCstandard). Response was assessed per iRECIST. Baseline parameters were compared between responders and nonresponders; subgroup analysis was conducted for clear cell RCC (ccRCC). Diagnostic performance was evaluated using the area under the curve (AUC). Immune infiltration was analyzed from RNA‑sequencing data, and correlations were assessed with the Spearman’s test. Among 58 patients (44 ccRCC), 33 (56.9

RATIONALE AND OBJECTIVES: This study aimed to develop and validate a magnetic resonance imaging (MRI)-based habitat radiomics model to predict recurrence-free survival (RFS) in patients with nonmetastatic clear cell renal cell carcinoma (ccRCC) after surgical resection. MATERIALS AND METHODS:A retrospective cohort of 630 patients with nonmetastatic ccRCC who underwent surgical resection at the First Medical Center of Chinese PLA General Hospital (2011-2019) was included. Preoperative T2-weighted imaging (T2WI) and contrast-enhanced corticomedullary phase (CP) MRI were used to cluster tumor voxels into homogeneous habitats via K-means algorithm based on signal intensity. Radiomic features were extracted from habitats; after feature selection, these features were integrated with clinicopathological indicators to build a Cox proportional hazards regression model. Model performance was assessed via receiver operating characteristic (ROC) curves, concordance index (C-index), calibration curves, and decision curve analysis (DCA). RESULTS:Three distinct tumor habitat regions were identified through clustering, from which 13 recurrence-related radiomic features were selected to construct a Habitat Signature (HS). Multivariate Cox regression analysis demonstrated that age (HR = 1.039, 95% CI: 1.016-1.063, P<0.001), sex (male vs female, HR = 2.608, 95% CI: 1.291-5.270, P=0.008), and pathological T stage (T3 vs T1, HR = 4.284, 95% CI: 1.997-9.193, P < 0.001) served as independent predictors of postoperative recurrence. Constructed by combining clinicopathological predictors with the HS score, the clinical-habitat combined model yielded AUC values for 3-year and 5-year postoperative recurrence prediction of 0.80/0.81 in the training set and 0.85/0.81 in the test set, along with C-indices of 0.80 and 0.81, respectively. CONCLUSION:The predictive model constructed by combining MRI-based HS score and clinical-pathological features has predictive value for recurrence of nonmetastatic ccRCC.

To evaluate whether MRI characteristics of septa and walls—specifically enhancement patterns during the corticomedullary (CP), nephrographic (NP), and excretory (EP) phases—can improve malignancy risk stratification in Bosniak III cystic renal masses (CRMs). This single-center, retrospective study analyzed 120 patients with Bosniak III cystic renal masses who underwent renal MRI between January 2009 and December 2021. The cohort included two subcategories: III-WS (enhancing thick wall/septa ≥ 4 mm; n = 22) and III-OP (enhancing irregular wall/septa or convex protrusion ≤ 3 mm; n = 98). All lesions were confirmed either by histopathology (115 CRMs) or ≥ 5 year stability (5 CRMs). Four radiologists (2 senior, 2 junior), blinded to clinical and pathological data, independently assessed septal and wall enhancement (obvious vs. non-obvious) across CP, NP) and EP, a consensus was reached through discussion. Interobserver agreement was evaluated using Conger’s kappa, while diagnostic performance (AUC, sensitivity, specificity, accuracy) of obvious enhancement was assessed via ROC analysis, with AUC comparisons performed using DeLong’s test. The study included 120 patients (mean age: 48 ± 11 years; 94 male), with 95 (79.2

OBJECTIVE:This study aims to develop a cascaded deep learning (DL) system based on multiparametric MRI to establish an automated pipeline for the segmentation and classification of small renal masses (SRMs). MATERIALS AND METHODS:A retrospective collection of SRM patients with pathologically confirmed from three institutions was conducted. MRI data from Institution 1 were randomly divided into a training set and an internal test set. Data from other institutions served as the external test set. A cascaded DL system was developed, incorporating automated segmentation and benign-malignant classification. Diagnostic performance was evaluated using receiver operating characteristic analysis and compared against three radiologists of varying experience. RESULTS:A total of 965 patients with SRM were included. Institution 1 contributed 888 cases, with 712 used for training and 176 as an internal test set; Institutions 2 and 3 provided 77 cases as an external test set. The optimal classification model using automated segmentation labels achieved AUCs of 0.936 and 0.788 on internal and external test sets, respectively. Performance was comparable to models using manual segmentation (internal: 0.936 vs. 0.944, P = 0.671; external: 0.788 vs. 0.832, P = 0.629). On the external test set, the model performed comparably to the senior radiologist, while it significantly outperformed the senior radiologist on the internal test set. The model significantly outperformed the junior radiologist on both test sets. This finding remained consistent in the subgroup of tumors smaller than 3 cm. CONCLUSION:The cascaded DL system demonstrated robust performance across multiple centers, enabling non-invasive and efficient discrimination of SRM malignancy, showing promise as a clinical support tool.

To develop and validate a machine learning (ML)-based pipeline for automated segmentation and classification of complicated cystic renal masses (cCRMs) on MRI. This multicenter retrospective study enrolled 275 patients (median age, 48 years; 85 females) with pathologically confirmed 275 cCRMs (203 malignant) who underwent renal MRI from January 2013 to December 2023. cCRMs from one institution were used as a training set (n = 215), while those from the other three institutions served as a test set (n = 60). 3D V-Net and random forest algorithms were employed for segmentation and classification, respectively. Segmentation and classification performance was evaluated using the Dice similarity coefficient (DSC) and the area under the curve (AUC), respectively. Two junior and two senior radiologists independently classified cCRMs in the test set into Bosniak categories II–IV based on the Bosniak classification, version 2019. In the test set, the ML pipeline achieved DSC of 0.718 for cCRMs (n = 60) on excretory phase images. Additionally, classification performance of the ML pipeline (AUC = 0.835, 95

OBJECTIVES:This study aims to develop an artificial intelligence (AI)-based automated segmentation method for small renal masses (SRMs) using multi-center, multi-scanner, multi-sequence MRI data. METHODS:MR images from 988 pathologically confirmed SRM patients from three different centers were retrospectively included. Segmentation networks were independently developed for each MRI sequence using deep learning techniques. A GE dataset of 733 patients from Center 1 was used for training and validation. A GE test set, consisting of internal (99 from Center 1) and external test sets (81 from Center 2 and 3), was created for evaluation. Furthermore, a non-GE generalization set, consisting of 75 patients from Center 2 and 3, was used to assess the generalization ability. The method's performance was evaluated in terms of detection rate and segmentation accuracy (Dice similarity coefficient [DSC]). Subgroup analysis and multiple linear regression were used for further exploration. RESULTS:Our method demonstrated promising results in the detection and segmentation of SRMs. All patients in the GE test set were correctly detected in at least one sequence. Our model achieved a median DSC of 0.769-0.855 across five MRI sequences and demonstrated reasonable generalization to non-GE scanners (median DSC range: 0.523-0.785). CONCLUSIONS:The implementation of automated segmentation achieved encouraging outcomes in both correct-detection rates and segmentation accuracy across a diverse cohort spanning multiple centers and scanners, suggesting its potential as a key component of future diagnostic pipelines for SRMs.

Renal tumor and inferior vena cava tumor thrombus (RT-IVCTT), as a unique model for investigating congestive renal injury, remains largely unexplored. Although chronic kidney disease significantly correlates with postoperative adverse events and long-term prognosis in both renal tumor and pan-cancer settings, there is no large-scale evidence specifically addressing RT-IVCTT. This study aimed to identify risk factors of preoperative renal dysfunction (pre-RD) and evaluate its contribution as a predictor for prognosis in patients with RT-IVCTT. Consecutive postoperative patients with RT-IVCTT and no evidence of distant metastasis (n = 208) were retrospectively enrolled between June 2013 and June 2023. Preoperative estimated glomerular filtration rate (pre-eGFR) (mL/min/1.73 m2) was classified into four categories: ≥ 120, 90–120, 60–90, and < 60, and clinical data were analyzed to determine a meaningful cutoff. Multivariate logistic and Cox regression models were employed to identify risk factors of pre-RD and prognosis in patients with RT-IVCTT, respectively. Pre-RD, defined as pre-eGFR < 90 mL/min/1.73 m2, was found in 59 patients (28.4

BACKGROUND:T1-hyperintensity in cystic renal masses (CRMs) complicates Bosniak classification assessment due to inherent signal interference from hemorrhage and proteinaceous content, potentially obscuring enhancement visibility. PURPOSE:To investigate whether MR subtraction imaging improves interobserver agreement and diagnostic performance in the Bosniak classification of T1-hyperintense CRMs. STUDY TYPE:Retrospective. POPULATION:A total of 139 consecutive patients (mean age, 50 ± 12 years; 97 males) with 141 T1-hyperintense CRMs were included, consisting of surgically confirmed 133 lesions and clinically diagnosed 8 benign CRMs that were stable during follow-up (≥ 5 years). FIELD STRENGTH/SEQUENCE:1.5/3 T. fat-saturated T2-weighted imaging, diffusion-weighted imaging, unenhanced, and triphasic dynamic contrast-enhanced T1-weighted imaging (T1WI). Subtraction images were generated automatically by subtracting unenhanced from triphasic contrast-enhanced T1WI. ASSESSMENT:Six radiologists (half less experienced) independently classified all T1-hyperintense CRMs using the Bosniak classification (v2019) in two sessions, with and without subtraction imaging. A 1-month washout period was implemented between sessions, and the order of cases was re-randomized. Interobserver agreement and diagnostic performance were evaluated in all experienced and less experienced readers. STATISTICAL TESTS:Weighted κ statistics assessed interobserver agreement. Diagnostic performance (the area under the curve [AUC], sensitivity, specificity) was compared using Delong and McNemar tests. Statistical significance was defined as p < 0.05. RESULTS:Subtraction imaging significantly improved interobserver agreement in all radiologists (weighted κ = 0.62 vs. 0.46), and less experienced radiologists (3-5 years of experience, weighted κ = 0.63 vs. 0.42), though not significantly among experienced radiologists (10-15 years of experience, weighted κ = 0.61 vs. 0.52; p = 0.051). Less experienced radiologists showed significantly higher AUC (0.865 vs. 0.804), sensitivity (88.9% vs. 75.5%), and specificity (88.2% vs. 72.5%) with MR subtraction imaging. DATA CONCLUSION:MR subtraction imaging may improve overall interobserver agreement in the Bosniak classification of T1-hyperintense CRMs. Furthermore, it could improve diagnostic accuracy and interobserver agreement among less experienced radiologists. EVIDENCE LEVEL:4. TECHNICAL EFFICACY:Stage 2.

To explore the influence of perirenal collateral (PRC) on the operation and prognosis of nonmetastatic renal cell carcinoma (RCC) and inferior vena cava (IVC) tumor thrombus. A total of 10,260 patients with RCC who underwent surgical resection were retrospectively collected. Two hundred and forty-one patients (median age, 56.0 years [IQR: 48.0, 63.0], 174 males) with pathologically confirmed nonmetastatic RCC and IVC tumor thrombus were analyzed. The presence or absence of abnormal PRC was determined by 3 observers on images. Relevant clinical data, details of the operation, postoperative complications, and pathological results were obtained and compared. The Kaplan-Meier method was used to estimate recurrence-free survival (RFS). A total of 185 patients had abnormal PRC on preoperative MR images. Patients with abnormal PRC had significantly larger tumor sizes (P < 0.001), were more likely to have clinical symptoms (P < 0.001), and had positive RBCs in urine (P < 0.001). During the operation, patients with abnormal PRC had a longer operative duration (P < 0.001), more blood loss (P < 0.001), and more red blood cell infusions (P = 0.002). Compared with patients without abnormal PRC, the proportion of open surgery was significantly higher in patients with abnormal PRC (44/185 vs. 5/56, P = 0.001). RFS was significantly shorter (P = 0.002) among patients with abnormal PRC. By multivariate Cox regression analysis, renal sinus fat invasion, perirenal fat invasion, tumor size, and abnormal PRC were independent predictors of RFS (all P < 0.05). In patients with nonmetastatic RCC and IVC tumor thrombus, abnormal PRC was associated with intraoperative adverse events and worse RFS. Evaluation of PRC through imaging could help develop treatment plans and provide prognostic information.

BACKGROUND:MRI assessment for extraprostatic extension (EPE) of prostate cancer (PCa) is challenging due to limited accuracy and interobserver agreement. PURPOSE:To develop an interpretable Tabular Prior-data Fitted Network (TabPFN)-based radiomics model to evaluate EPE using MRI and explore its integration with radiologists' assessments. STUDY TYPE:Retrospective. POPULATION:Five hundred and thirteen consecutive patients who underwent radical prostatectomy. Four hundred and eleven patients from center 1 (mean age 67 ± 7 years) formed training (287 patients) and internal test (124 patients) sets, and 102 patients from center 2 (mean age 66 ± 6 years) were assigned as an external test set. FIELD STRENGTH/SEQUENCE:Three Tesla, fast spin echo T2-weighted imaging (T2WI) and diffusion-weighted imaging using single-shot echo planar imaging. ASSESSMENT:Radiomics features were extracted from T2WI and apparent diffusion coefficient maps, and the TabRadiomics model was developed using TabPFN. Three machine learning models served as baseline comparisons: support vector machine, random forest, and categorical boosting. Two radiologists (with > 1500 and > 500 prostate MRI interpretations, respectively) independently evaluated EPE grade on MRI. Artificial intelligence (AI)-modified EPE grading algorithms incorporating the TabRadiomics model with radiologists' interpretations of curvilinear contact length and frank EPE were simulated. STATISTICAL TESTS:Receiver operating characteristic curve (AUC), Delong test, and McNemar test. p < 0.05 was considered significant. RESULTS:The TabRadiomics model performed comparably to machine learning models in both internal and external tests, with AUCs of 0.806 (95% CI, 0.727-0.884) and 0.842 (95% CI, 0.770-0.912), respectively. AI-modified algorithms showed significantly higher accuracies compared with the less experienced reader in internal testing, with up to 34.7% of interpretations requiring no radiologist input. However, no difference was observed in both readers in the external test set. DATA CONCLUSIONS:The TabRadiomics model demonstrated high performance in EPE assessment and may improve clinical assessment in PCa. EVIDENCE LEVEL:4. TECHNICAL EFFICACY:Stage 2.

Rationale and Objective: Accurate differentiation between benign and malignant cystic renal masses (CRMs) is challenging in clinical practice. This study aimed to develop MRI-based machine learning models for differentiating between benign and malignant CRMs and compare the best-performing model with the Bosniak classification, version 2019 (BC, version 2019). Methods: Between 2009 and 2021, consecutive surgery-proven CRM patients with renal MRI were enrolled in this multicenter study. Models were constructed to differentiate between benign and malignant CRMs using logistic regression (LR), random forest (RF), and support vector machine (SVM) algorithms, respectively. Meanwhile, two radiologists classified CRMs into I-IV categories according to the BC, version 2019 in consensus in the test set. A subgroup analysis was conducted to investigate the performance of the best- performing model in complicated CRMs (II-IV lesions in the test set). The performances of models and BC, version 2019 were evaluated using the area under the receiver operating characteristic curve (AUC). Performance was statistically compared between the best- performing model and the BC, version 2019. Results: 278 and 48 patients were assigned to the training and test sets, respectively. In the test set, the AUC and accuracy of the LR model, the RF model, the SVM model, and the BC, version 2019 were 0.884 and 75.0%, 0.907 and 83.3%, 0.814 and 72.9%, and 0.893 and 81.2%, respectively. Neither the AUC nor the accuracy of the RF model that performed best were significantly different from the BC, version 2019 (P = 0.780, P = 0.065). The RF model achieved an AUC and accuracy of 0.880 and 81.0% in complicated CRMs. Conclusions: The MRI-based RF model can accurately differentiate between benign and malignant CRMs with comparable perfor- mance to the BC, version 2019, and has good performance in complicated CRMs, which may facilitate treatment decision-making and is less affected by interobserver disagreements.

Rationale and Objectives To improve the diagnostic recognition of papillary renal neoplasm with reverse polarity (PRNRP) through comprehensive analysis of computed tomography (CT) and magnetic resonance imaging (MRI) findings. Materials and Methods A retrospective multi-center study was conducted on patients with pathologically confirmed PRNRPs from 2019 to 2024, encompassing six institutions. Clinical and pathological data were meticulously documented. Preoperative CT (n=23) and MRI (n=9) features were independently evaluated in consensus by two genitourinary radiologists, focusing on tumor location, morphologic features, attenuation, signal intensity, and enhancement patterns. Postoperative outcomes were assessed through medical record review or telephone follow-up. Results The study cohort comprised 26 patients (mean age 62±12 years, 13 men) with 26 well-defined PRNRPs (mean diameter 2.2±1.0 cm) were included. 15 cases (58%) were situated in the right kidney, 18(69%) were exophytic, and 23(88%) were quasi-spherical. A pseudocapsule was identified in eight cases (89%) on MRI. All cases demonstrated iso- or slight hyperattenuation (43.1±13.6 HU) on non-contrast CT, hypointensity on T2-weighted imaging (T2WI), and mild diffusion restriction on diffusion-weighted imaging (DWI). All cases exhibited mild or moderate enhancement in the corticomedullary phase, followed by progressive enhancement in the nephrographic and excretory phases. Concomitant renal cysts were found in 17 cases (65%). All cases showed no evidence of recurrence or metastasis. Conclusion PRNRP typically presents as a small hypovascular renal mass, and characterized by a pseudocapsule, iso- or slight hyperattenuation on non-contrast CT, heterogeneous T2WI hypointensity, and mild diffusion restriction on DWI.

To differentiate mixed epithelial and stromal tumor family (MESTF) of the kidney from predominantly cystic renal cell carcinoma (RCC) using the magnetic resonance imaging (MRI)-based Bosniak classification system version 2019 (v2019). The study included 36 consecutive patients with MESTF and 77 with predominantly cystic RCC who underwent preoperative renal MRI. One radiologist evaluated and documented the clinical and MRI characteristics (age, sex, laterality, R.E.N.A.L. Nephrometry Score [RNS], surgical approach, the signal intensity on T2-weighted imaging, restricted diffusion and enhancement features in corticomedullary phase). Blinded to clinical and pathological information, another two radiologists independently evaluated Bosniak category of all masses. Interobserver agreement based on Bosniak classification system v2019 was measured by the weighted Cohen/Conger’s Kappa coefficient. Furthermore, predominantly cystic RCCs and MESTFs were divided into low (categories I, II, and IIF) and high-class (categories III, and IV) tumors. The independent sample t test (Mann–Whitney U test) or Pearson Chi-square test (Fisher’s exact probability test) was utilized to compare clinical and imaging characteristics between MESTFs and predominantly cystic RCCs. The performance of the Bosniak classification system v2019 in distinguishing MESTF from predominantly cystic RCC was investigated via receiver operating characteristic curve analysis. MESTF and predominantly cystic RCC groups significantly differed in terms of age, lesion size, RNS, restricted diffusion, and obvious enhancement in corticomedullary phase, but not sex, laterality, surgical approach, and the signal intensity on T2WI. Interobserver agreement was substantially based on the Bosniak classification system v2019. There were 24 low-class tumors and 12 high-class tumors in the MESTF group. Meanwhile, 13 low-class tumors and 64 high-class tumors were observed in the predominantly cystic RCC group. The distribution of low- or high-class tumors significantly differed between the MESTF and predominantly cystic RCC groups. Bosniak classification system v2019 had excellent discrimination (cutoff value = category III), and an area under curve value was 0.81; accuracy, 80.5

Abstract Objective To assess the utility of multiparametric MRI and clinical indicators in distinguishing nuclear grade and survival of clear cell renal cell carcinoma (ccRCC) complicated with venous tumor thrombus (VTT). Materials and methods This study included 105 and 27 patients in the training and test sets, respectively. Preoperative MRI, including intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI), was performed. Renal lesions were evaluated for IVIM-DWI metrics and conventional MRI features. All the patients had postoperative histologically proven ccRCC and VTT. An expert uropathologist reviewed all specimens to confirm the nuclear grade of the World Health Organization/ International Society of Urological Pathology (WHO/ISUP) of the tumor. Univariate and multivariable logistic regression analyses were used to select the preoperative imaging features and clinical indicators. The predictive ability of the logistic regression model was assessed using receiver operating characteristic (ROC) analysis. Survival curves were plotted using the Kaplan–Meier method. Results High WHO/ISUP nuclear grade was confirmed in 69 of 105 patients (65.7%) in the training set and 19 of 27 patients (70.4%) in the test set, respectively (P = 0.647). Dp_ROI_Low, tumor size, serum albumin, platelet count, and lymphocyte count were independently related to high WHO/ISUP nuclear grade in the training set. The model identified high WHO/ISUP nuclear grade well, with an AUC of 0.817 (95% confidence interval [CI]: 0.735–0.899), a sensitivity of 70.0%, and a specificity of 77.8% in the training set. In the independent test set, the model demonstrated an AUC of 0.766 (95% CI, 0.567–0.966), a sensitivity of 79.0%, and a specificity of 75.0%. Kaplan–Meier analysis showed that the predicted high WHO/ISUP nuclear grade group had poorer progression-free survival than the low WHO/ISUP nuclear grade group in both the training and test sets (P = 0.001 and P = 0.021). Conclusions IVIM-DWI-derived parameters and clinical indicators can be used to differentiate nuclear grades and predict progression-free survival of ccRCC and VTT.

Background Venous tumor thrombus (VTT) consistency of renal cell carcinoma (RCC) is an important consideration in nephrectomy plus thrombectomy. However, evaluation of VTT consistency through preoperative MR imaging is lacking. Purpose To evaluate VTT consistency of RCC through intravoxel incoherent motion‐diffusion weighted imaging (IVIM‐DWI) derived parameters ( D t , D p , f , and ADC) and the apparent diffusion coefficient (ADC) value. Study Type Retrospective. Population One hundred and nineteen patients (aged 55.8 ± 11.5 years, 85 male) with histologically‐proven RCC and VTT who underwent radical resection. Field Strength/Sequences 3.0‐T; two‐dimensional single‐shot diffusion‐weighted echo planar imaging sequence at 9 b ‐values (0–800 s/mm 2 ). Assessment IVIM parameters and ADC values of the primary tumor and the VTT were calculated. The VTT consistency (friable vs. solid) was determined through intraoperative findings of two urologists. The accuracy of VTT consistency classification based on the individual IVIM parameters of primary tumors and of VTT, and based on models combining parameters, was assessed. Type of operation, intra‐operative blood loss, and operation length were recorded. Statistical Tests Shapiro–Wilk test; Mann–Whitney U test; Student's t ‐test; Chi‐square test; Receiver operating characteristic (ROC) analysis. Statistical significance level was P < 0.05. Results Of the enrolled 119 patients, 33 patients (27.7%) had friable VTT. Patients with friable VTT were significantly more likely to experience open surgery, have significantly more intraoperative blood loss, and significantly longer operative duration. The area under the ROC curve (AUC) values of D t of the primary tumor and VTT in classifying VTT consistency were 0.758 (95% CI 0.671–0.832) and 0.712 (95% CI 0.622–0.792), respectively. The AUC value of the model combining D p and D t of VTT was 0.800 (95% CI 0.717–0.868). Furthermore, the AUC of the model combining D p and D t of VTT and D t of the primary tumor was 0.886 (95% CI 0.814–0.937). Conclusion IVIM‐derived parameters had the potential to predict VTT consistency of RCC. Evidence Level: 3 Technical Efficacy: Stage 2

Objective:To investigate the clinical and MRI features of the mixed epithelial and stromal tumor family (MESTF) of the kidney.Methods:From January 2009 to September 2021, 42 patients with pathologically-proven MESTF from the First Medical Center of Chinese PLA General Hospital were collected in this retrospective study. Clinical information, MRI features, and pathological results were documented. According to the Bosniak classification (BC) version 2019, all MESTFs were divided into cystic MESTFs (36 cases) and solid-cystic MESTFs (6 cases). The R.E.N.A.L. nephrometry score (RNS), lesion size, laterality, location, margin, shape, growth pattern, presence of protruding into renal sinus, hemorrhage, and enhancement pattern were evaluated and documented. Based on BC versions 2005 and 2019, all the cystic MESTFs were assessed and divided into low (Ⅰ, Ⅱ, ⅡF) and high (Ⅲ, Ⅳ) grades. The independent sample t test or Mann-Whitney U test were performed to compare age, RNS, and lesion size between cystic MESTFs and solid-cystic MESTFs. Pearson χ 2 test, continuity-adjusted χ 2 test or Fisher exact probability test were utilized to evaluated the differences of clinical and MRI features and the distribution of low or high grades in two versions of BC. Results:Forty-two MESTFs were unilateral and solitary masses, 25 males and 17 females, with a mean age of (41±13) years old. Compared to solid-cystic MESTFs, cystic MESTFs were prone to demonstrate endophytic growth pattern (χ 2=17.77, P<0.001), and no significant differences in other clinical and MRI features were observed between cystic and solid-cystic MESTFs (all P>0.05). There were 7 low-grade and 29 high-grade tumors in the BC version 2005, respectively. Meanwhile, 24 low-grade and 12 high-grade tumors in the BC version 2019, respectively. The distribution of low or high-grade tumors in the two versions of BC had a statistically significant difference (χ 2=16.37, P<0.001). Conclusion:MESTFs demonstrated middle-age onset and no gender predilection. Cystic MESTFs are more likely to exhibit endophytic growth pattern with low-grade classification in BC system version 2019.

Objective:To analyze the MRI characteristics of surgical resected renal angiomyolipoma (AML) smaller than 4 cm.Methods:A total of 112 patients with surgical pathology confirmed renal AML of which the maximum diameter was smaller than 4 cm were analyzed retrospectively in the First Medical Centre, Chinese PLA General Hospital from January 2014 to November 2020, 5 of which were epithelioid angiomyolipoma (EAML) patients. According to the presence or absence of visible fat in lesions on MRI, the lesions were divided into AML with fat group and AML without visible fat (AML wovf) group. The MRI features were evaluated, including maximum diameter, location, growth pattern, shape, beak sign, angular interface with renal parenchyma, pseudo-capsule, hemorrhage, cystic degeneration, coagulative necrosis, flowing void in the tumor, signal intensity and homogeneity on T 2WI and diffusion weighter imaging (DWI), the peak enhanced phase. The differences of maximum diameter of AML with fat and AML wovf were analyzed using Mann-Whitney U test, and the differences of MRI features were analyzed using χ 2 test or Fisher′s exact probability test. Results:There were 123 lesions found in 112 patients, and 96 lesions contained fat and 27 lesions were AML wovf. 82 lesions showed round and round-like shapes, 112 lesions showed exophytic growth pattern, 71 lesions with peak enhancement in corticomedullary phase. And the numbers of lesions with angular interface with renal parenchyma, beak sign, cystic degeneration, pseudo-capsule, hemorrhage were 30, 49, 1, 1, 1, respectively. There was no coagulative necrosis in all lesions. Compared with AML with fat, AML wovf was single lesion. The diameters of AML with fat and AML wovf were 2.5 (1.7, 3.5) and 1.8 (1.4, 2.3) cm respectively, with statistically significant difference ( Z=-2.80, P=0.005). In the AML with fat and AML wovf, 65 and 12 cases were heterogeneous in T 2WI, 44 and 5 lesions showed beak sign, 26 and 4 lesions showed angular interface with renal parenchyma, 57 and 10 cases were heterogeneous in DWI. And there were 5 and 6 lesions showed the endophytic, 44 and 8 lesions showed partly exophytic, 47 and 13 lesions showed exophytic in patterns of tumor growth respectively. The beak sign, homogeneous in T 2WI and DWI, patterns of tumor growth showed statistical differences in AML with fat and AML wovf (all P<0.05), and there was no significant difference in other features ( P>0.05). A total of 5 EAML patients were with 8 lesions. One patient had 4 lesions with fat, other patients had single lesion in which 2 lesions with fat, 2 lesions without visible fat. One lesion without visible fat showed hemorrhage. Conclusions:Surgical resected AML smaller than 4 cm is often exophytic round and round-like, enhanced in corticomedullary phase, showing angular interface with renal parenchyma and beak sign, with rare cystic degeneration, pseudo-capsule, hemorrhage and improbable coagulation necrosis. AML wovf is single smaller lesion which often shows endophytic growth pattern, and beak sign is infrequent. EAML seems to be present in two modes, multiple lesions with fat and AML wovf with hemorrhage.

Objective:To investigate the value of MR subtraction images in improving the interobserver agreement for Bosniak Ⅱ, ⅡF, and Ⅲ cystic renal masses (CRMs) with Bosniak classification version 2019.Methods:From January 2009 to August 2020, 323 patients (335 CRMs) with surgical pathology results and complete preoperative MRI examination (T 2WI, T 1WI precontrast images and enhanced MRI in corticomedullary, nephrographic, and excretory phases) were retrospectively collected in the First Medical Center of PLA General Hospital. The CRMs of Bosniak Ⅱ, ⅡF, and Ⅲ were selected and classified by 2 experienced genitourinary radiologists according to the Bosniak classification version 2019. The "Subtraction" function in the American GE ADW 4.4 workstation was used to perform subtraction images reconstruction on the enhanced images in the corticomedullary, nephrographic, and excretory phases. Blinded to pathologic information, the other 2 radiologists independently classified the enrolled CRM cases with and without subtraction MR images, respectively, with an interval of 1 month. Ultimately, by using weighted Kappa value, interobserver agreement was evaluated, and the differences in weighted Kappa value were compared using the Gwet coefficient. Results:A total of 187 patients with 187 CRMs were enrolled in the study. The results of the classification of Bosniak Ⅱ, ⅡF, and Ⅲ CRMs categorized by 2 radiologists without and with subtraction images showed that 119 and 141 cases were consistent, and 68 and 46 were inconsistent, respectively. The weighted Kappa value for interobserver agreement among two radiologists without and with subtraction MR images was 0.60 (95%CI 0.53-0.68) and 0.73 (95%CI 0.66-0.80), respectively. The interobserver agreement was higher with subtraction images than that without subtraction images ( t=-2.56, P=0.011). Conclusion:According to the MRI criteria of Bosniak classification version 2019, the interobserver agreement for Bosniak Ⅱ, ⅡF, and Ⅲ CRMs could be improved using subtraction MR images, which may facilitate the popularization and application of Bosniak classification version 2019.