Total Tumor ADC Histogram Analysis: A New Tool for Predicting High-Risk Cytogenetic Abnormalities in Multiple Myeloma Patients.

RATIONALE AND OBJECTIVES: We explored the feasibility of using total tumor apparent diffusion coefficient (ttADC) histogram parameters to predict high-risk cytogenetic abnormalities (HRCA) in patients with multiple myeloma (MM) and compared the performance of an image prediction model based on these parameters with that of a combined prediction model based on these parameters and clinical indicators. METHODS: We retrospectively analyzed the parameters of the ttADC histogram based on whole-body diffusion-weighted images(WB-DWI) and clinical indicators in 92 patients with MM. The patients were divided into HRCA and non-HRCA groups according to the results of the fluorescence in situ hybridization. Logistic regression analysis was used to construct the image prediction and combined prediction models. The area under the curve (AUC) of the receiver operating characteristic (ROC) curve was used to evaluate the performance of the models to identify HRCA. The DeLong test was used to compare the AUC differences of each prediction model. RESULTS: Logistic regression analysis results revealed that the ttADC histogram parameter, ttADC entropy < 7.959 (OR: 39.167; 95% confidence interval [CI]: 3.891-394.208; P < 0.05), was an independent risk factor for HRCA. The image prediction model consisted of ttADC entropy and ttADC SD. The combined prediction model included ttADC entropy along with patient clinical indicators such as biological sex and M protein percentage. The AUCs of the image prediction and combined prediction models were 0.739 and 0.811, respectively (P < .05). The image prediction model showed a sensitivity of 73.9% and a specificity of 68.1%. The combined prediction model showed 82.6% sensitivity and 72.5% specificity. CONCLUSIONS: Using ttADC histogram parameters based on WB-DWI images to predict HRCA in patients with MM is feasible, and combining ttADC parameters with clinical indicators can achieve better predictive performance.

[Applications of Artificial Intelligence in Pancreatic Cystic Lesion Imaging].

As the detection rate of pancreatic cystic lesions(PCL)increases,artificial intelligence(AI)has made breakthroughs in the imaging workflow of PCL,including image post-processing,lesion detection,segmentation,diagnosis and differential diagnosis.AI-based image post-processing can optimize the quality of medical images and AI-assisted models for lesion detection,segmentation,diagnosis and differential diagnosis significantly enhance the work efficiency of radiologists.This article reviews the application progress of AI in PCL imaging and provides prospects for future research directions.

Quantitative MR imaging biomarkers for distinguishing inflammatory pancreatic mass and pancreatic cancer-a systematic review and meta-analysis.

OBJECTIVES: To evaluate the diagnostic performance of quantitative magnetic resonance (MR) imaging biomarkers in distinguishing between inflammatory pancreatic masses (IPM) and pancreatic cancer (PC). METHODS: A literature search was conducted using PubMed, Embase, the Cochrane Library, and Web of Science through August 2023. Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2) was used to evaluate the risk of bias and applicability of the studies. The pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio were calculated using the DerSimonian-Laird method. Univariate meta-regression analysis was used to identify the potential factors of heterogeneity. RESULTS: Twenty-four studies were included in this meta-analysis. The two main types of IPM, mass-forming pancreatitis (MFP) and autoimmune pancreatitis (AIP), differ in their apparent diffusion coefficient (ADC) values. Compared with PC, the ADC value was higher in MFP but lower in AIP. The pooled sensitivity/specificity of ADC were 0.80/0.85 for distinguishing MFP from PC and 0.82/0.84 for distinguishing AIP from PC. The pooled sensitivity/specificity for the maximal diameter of the upstream main pancreatic duct (dMPD) was 0.86/0.74, with a cutoff of dMPD ≤ 4 mm, and 0.97/0.52, with a cutoff of dMPD ≤ 5 mm. The pooled sensitivity/specificity for perfusion fraction (f) was 0.82/0.68, and 0.82/0.77 for mass stiffness values. CONCLUSIONS: Quantitative MR imaging biomarkers are useful in distinguishing between IPM and PC. ADC values differ between MFP and AIP, and they should be separated for consideration in future studies. CLINICAL RELEVANCE STATEMENT: Quantitative MR parameters could serve as non-invasive imaging biomarkers for differentiating malignant pancreatic neoplasms from inflammatory masses of the pancreas, and hence help to avoid unnecessary surgery. KEY POINTS: • Several quantitative MR imaging biomarkers performed well in differential diagnosis between inflammatory pancreatic mass and pancreatic cancer. • The ADC value could discern pancreatic cancer from mass-forming pancreatitis or autoimmune pancreatitis, if the two inflammatory mass types are not combined. • The diameter of main pancreatic duct had the highest specificity for differentiating autoimmune pancreatitis from pancreatic cancer.

CARE-radiology statement explanation and elaboration: reporting guideline for radiological case reports.

Despite the increasing number of radiological case reports, the majority lack a standardised methodology of writing and reporting. We therefore develop a reporting guideline for radiological case reports based on the CAse REport (CARE) statement. We established a multidisciplinary group of experts, comprising 40 radiologists, methodologists, journal editors and researchers, to develop a reporting guideline for radiological case reports according to the methodology recommended by the Enhancing the QUAlity and Transparency Of health Research network. The Delphi panel was requested to evaluate the significance of a list of elements for potential inclusion in a guideline for reporting mediation analyses. By reviewing the reporting guidelines and through discussion, we initially drafted 46 potential items. Following a Delphi survey and discussion, the final CARE-radiology checklist is comprised of 38 items in 16 domains. CARE-radiology is a comprehensive reporting guideline for radiological case reports developed using a rigorous methodology. We hope that compliance with CARE-radiology will help in the future to improve the completeness and quality of case reports in radiology.

Association of lower extremity peripheral arterial disease with quantitative muscle features from computed tomography angiography.

OBJECTIVES: To explore the association between lower extremity muscle features from CTA and peripheral arterial disease (PAD) severity using digital subtraction angiography (DSA) as reference standard. METHODS: Informed consent was waived for this Institutional Review Board approved retrospective study. PAD patients were recruited from July 2016 to September 2020. Two radiologists evaluated PAD severity on DSA and CTA using runoff score. The patients were divided into two groups: mild PAD (DSA score ≤ 7) vs. severe PAD (DSA score > 7). After segmenting lower extremity muscles from CTA, 95 features were extracted for univariable analysis, logistic regression model (LRM) analysis, and sub-dataset analysis (PAD prediction based on only part of the images). AUC of CTA score and LRMs for PAD prediction were calculated. Features were analyzed using Student's t test and chi-squared test. p < 0.05 was considered statistically significant. RESULTS: A total of 56 patients (69 ± 11 years; 38 men) with 56 lower legs were enrolled in this study. The lower leg muscles of mild PAD group (36 patients) showed higher CT values (44.6 vs. 39.5, p < 0.001) with smaller dispersion (35.6 vs. 41.0, p < 0.001) than the severe group (20 patients). The AUC of CTA score, LRM-I (constructed with muscle features), and LRM-II (constructed with muscle features and CTA score) for PAD severity prediction were 0.81, 0.84, and 0.89, respectively. The highest predictive performance was observed in the image subset of the middle and inferior segments of lower extremity (LRM-I, 0.83; LRM-II, 0.90). CONCLUSIONS: Lower extremity muscle features are associated with PAD severity and can be used for PAD prediction. CRITICAL RELEVANCE STATEMENT: Quantitative image features of lower extremity muscles are associated with the degree of lower leg arterial stenosis/occlusion and can be a beneficial supplement to the current imaging methods of vascular stenosis evaluation for the prediction of peripheral arterial disease severity. KEY POINTS: • Compared with severe PAD, lower leg muscles of mild PAD showed higher CT values (39.5 vs. 44.6, p < 0.001). • Models developed with muscle CT features had AUC = 0.89 for predicting PAD. • PAD severity prediction can be realized through the middle and inferior segment of images (AUC = 0.90).

Damage evaluation of craniofacial localized scleroderma using magnetic resonance imaging.

Background: Localized scleroderma (LoS) is an autoimmune disease in which craniofacial lesions can cause severe facial deformities with brain involvement. Objective evaluation of craniofacial LoS is challenging. Magnetic resonance imaging (MRI) may be used as a damage assessment tool. This study aimed to analyze the tissue involvement of craniofacial LoS based on MRI and evaluate MRI for craniofacial LoS assessment. Methods: This cross-sectional study included patients with craniofacial LoS from September 2021 to August 2022 in Peking Union Medical College Hospital. Patients who were clinically assessed in a stable phase were enrolled; patients with previous surgical treatment or contraindications to MRI were excluded. Participants underwent clinical, MRI, and ultrasound assessments. MRI was compared with ultrasound by correlation analysis and Bland-Altman analysis. The involvement of different tissues and different facial subunits was compared. The accumulated soft tissue atrophy index (ASTAI) was compared with clinical scores by correlation analysis. Results: A total of 28 patients were included (13 female; mean age, 18 years). MRI showed a good correlation and agreement with ultrasound (r=0.916, P<0.001). In different facial subunits, a significant negative correlation between the forehead and chin was found (r=-0.593, P=0.001). The ASTAI correlated well with the facial LoS damage index (r=0.580, P=0.001) and the Peking Union Medical College LoS facial aesthetic index (PUMC LoSFAI) (r=0.921, P<0.001). A total of 38.6% of clinical scores were inaccurate based on MRI. Neurological changes were found in one patient. Conclusions: MRI can reliably quantify damage in craniofacial LoS, and may serve as a useful and objective tool for overall craniofacial LoS evaluation.

Magnetic-optical dual-modality imaging monitoring chemotherapy efficacy of pancreatic ductal adenocarcinoma with a low-dose fibronectin-targeting Gd-based contrast agent.

PURPOSE: Pancreatic ductal adenocarcinoma (PDAC) is a lethal hypovascular tumor surrounded by dense fibrosis. Albumin-bound paclitaxel and gemcitabine (AG) chemotherapy is the mainstay of PDAC treatment through depleting peritumoral fibrosis and killing tumor cells; however, it remains challenging due to the lack of a noninvasive imaging method evaluating fibrotic changes during AG chemotherapy. In this study, we developed a dual-modality imaging platform that enables noninvasive, dynamic, and quantitative assessment of chemotherapy-induced fibrotic changes through near-infrared fluorescence molecular imaging (FMI) and magnetic resonance imaging (MRI) using an extradomain B fibronectin (EDB-FN)-targeted imaging probe (ZD2-Gd-DOTA-Cy7). METHODS: The ZD2-Gd-DOTA-Cy7 probe was constructed by conjugating a peptide (Cys-TVRTSAD) to Gd-DOTA and the near-infrared dye Cy7. PDAC murine xenograft models were intravenously injected with ZD2-Gd-DOTA-Cy7 at a Gd concentration of 0.05 mmol/kg or free Cy7 and Gd-DOTA as control. The normalized tumor background ratio (TBR) on FMI and the T1 reduction ratio on MRI were quantitatively analyzed. For models receiving AG chemotherapy or saline, MRI/FMI was performed before and after treatment. Histological analyses were performed for validation. RESULTS: The ZD2-Gd-DOTA-Cy7 concentration showed a linear correlation with the fluorescence intensity and T1 relaxation time in vitro. The optimal imaging time was 30 min after injection of the ZD2-Gd-DOTA-Cy7 (0.05 mmol/kg), only half of the clinic dosage of gadolinium. Additionally, ZD2-Gd-DOTA-Cy7 generated a 1.44-fold and 1.90-fold robust contrast enhancement compared with Cy7 (P < 0.05) and Gd-DOTA (P < 0.05), respectively. For AG chemotherapy monitoring, the T1 reduction ratio and normalized TBR in the fibrotic tumor areas were significantly increased by 1.99-fold (P < 0.05) and 1.78-fold (P < 0.05), respectively, in the control group compared with those in the AG group. CONCLUSION: MRI/FMI with a low dose of ZD2-Gd-DOTA-Cy7 enables sensitive imaging of PDAC and the quantitative assessment of fibrotic changes during AG chemotherapy, which shows potential clinical applications for precise diagnosis, post-treatment monitoring, and disease management.

Application of magnetic resonance imaging radiomics in endometrial cancer: a systematic review and meta-analysis.

PURPOSE: We aimed to systematically assess the methodological quality and clinical potential application of published magnetic resonance imaging (MRI)-based radiomics studies about endometrial cancer (EC). METHODS: Studies of EC radiomics analyses published between 1 January 2000 and 19 March 2023 were extracted, and their methodological quality was evaluated using the radiomics quality score (RQS) and Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Pairwise correlation analyses and separate meta-analyses of studies exploring differential diagnoses and risk prediction were also performed. RESULTS: Forty-five studies involving 3 aims were included. The mean RQS was 13.77 (range: 9-22.5); publication bias was observed in the areas of 'index test' and 'flow and timing'. A high RQS was significantly associated with therapy selection-aimed studies, low QUADAS-2 risk, recent publication year, and high-performance metrics. Raw data from 6 differential diagnosis and 34 risk prediction models were subjected to meta-analysis, revealing diagnostic odds ratios of 23.81 (95% confidence interval [CI] 8.48-66.83) and 18.23 (95% CI 13.68-24.29), respectively. CONCLUSION: The methodological quality of radiomics studies involving patients with EC is unsatisfactory. However, MRI-based radiomics analyses showed promising utility in terms of differential diagnosis and risk prediction.

Tomoelastography and Pancreatic Extracellular Volume Fraction Derived From MRI for Predicting Clinically Relevant Postoperative Pancreatic Fistula.

BACKGROUND: Pancreatic stiffness and extracellular volume fraction (ECV) are potential imaging biomarkers for pancreatic fibrosis. Clinically relevant postoperative fistula (CR-POPF) is one of the most severe complications after pancreaticoduodenectomy. Which imaging biomarker performs better for predicting the risk of CR-POPF remains unknown. PURPOSE: To evaluate the diagnostic performance of ECV and tomoelastography-derived pancreatic stiffness for predicting the risk of CR-POPF in patients undergoing pancreaticoduodenectomy. STUDY TYPE: Prospective. POPULATION: Eighty patients who underwent multiparametric pancreatic MRI before pancreaticoduodenectomy, among whom 16 developed CR-POPF and 64 did not. FIELD STRENGTH/SEQUENCE: 3 T/tomoelastography and precontrast and postcontrast T1 mapping of the pancreas. ASSESSMENT: Pancreatic stiffness was measured on the tomographic c-map, and pancreatic ECV was calculated from precontrast and postcontrast T1 maps. Pancreatic stiffness and ECV were compared with histological fibrosis grading (F0-F3). The optimal cutoff values for predicting CR-POPF were determined, and the correlation between CR-POPF and imaging parameters was evaluated. STATISTICAL TESTS: The Spearman's rank correlation and multivariate linear regression analysis was conducted. The receiver operating characteristic curve analysis and logistic regression analysis was performed. A double-sided P < 0.05 indicated a statistically significant difference. RESULTS: Pancreatic stiffness and ECV both showed a significantly positive correlation with histological pancreatic fibrosis (r = 0.73 and 0.56, respectively). Patients with advanced pancreatic fibrosis had significantly higher pancreatic stiffness and ECV compared to those with no/mild fibrosis. Pancreatic stiffness and ECV were also correlated with each other (r = 0.58). Lower pancreatic stiffness (<1.38 m/sec), lower ECV (<0.28), nondilated main pancreatic duct (<3 mm) and pathological diagnosis other than pancreatic ductal adenocarcinoma were associated with higher risk of CR-POPF at univariate analysis, and pancreatic stiffness was independently associated with CR-POPF at multivariate analysis (odds ratio: 18.59, 95% confidence interval: 4.45, 77.69). DATA CONCLUSION: Pancreatic stiffness and ECV were associated with histological fibrosis grading, and pancreatic stiffness was an independent predictor for CR-POPF. LEVEL OF EVIDENCE: 1 TECHNICAL EFFICACY STAGE: 5.

Intravoxel incoherent motion imaging to assess the acute effects of moderate-intensity continuous training and high-intensity interval training on thigh muscles.

This study investigated the use of intravoxel incoherent motion imaging (IVIM) to compare skeletal muscle perfusion during and after high-intensity interval training (HIIT) and moderate-intensity continuous training (MICT) to determine the impact on fat oxidation outcomes. Twenty overweight volunteers were recruited for the study. Each participant received one HIIT intervention and one MICT intervention using a cycling ergometer. Participants underwent a magnetic resonance imaging scan before, immediately after, and 1 and 2 h after each intervention. The IVIM parameters (D, fD*) of the rectus femoris, vastus lateralis, and biceps femoris long head were obtained. Changes in IVIM parameters of these muscles after both exercise interventions were compared using a two-factor repeated measures analysis of variance. In the rectus femoris, the fD* increased immediately after exercise intervention (d = 0.69 × 10-3  mm2 /s, p < 0.0083) and 2 h after exercise intervention (d = 0.64 × 10-3  mm2 /s, p < 0.0083) compared with before exercise. The increase in the fD* in the HIIT group was greater than that in the MICT group (d = 0.32, p = 0.023). In the vastus lateralis, the fD* increased immediately after the exercise intervention (d = 0.53 × 10-3  mm2 /s, p < 0.001) and returned to the pre-exercise level 1 h after exercising. The increase in the fD* in the HIIT group was lower than that in the MICT group (d = -0.21, p = 0.015). For the biceps femoris long head, the fD* was not significantly different between the two exercise interventions before and after exercise. Furthermore, the fD* 60 min after the HIIT intervention correlated with maximal oxygen consumption (VO2max), whereas fD* immediately after the MICT intervention correlated with VO2max. In summary, IVIM parameters can be used to evaluate differences in muscle perfusion between HIIT and MICT, and show a correlation with VO2max.

Treatment Response Assessment in Multiple Myeloma: Histogram Analysis of Total Tumor Apparent Diffusion Coefficient based on Whole-body Diffusion-weighted MR Imaging.

BACKGROUND: The International Myeloma Working Group (IMWG) consensus criteria for response assessment in multiple myeloma (MM) has methodological limitations. Whole-body diffusion-weighted imaging (DWI) with apparent diffusion coefficient (ADC) histogram analysis may be complementary to response assessment of MM. PURPOSE: To explore the role of histogram analysis of the ADC based on the total tumor volume (ttADC) in response assessment in patients with newly diagnosed MM (NDMM). STUDY TYPE: Retrospective. POPULATION: Thirty-six patients with NDMM. FIELD STRENGTH/SEQUENCE: 3.0T/single-shot DWI echo planar imaging (EPI) sequence with an integrated slice-by-slice shimming (iShim) technique. ASSESSMENT: Baseline (median: 1 day before treatment) and post-treatment (median: five cycles of therapy) whole-body DWI were analyzed. A region of interest (ROI) containing lesions on every section of baseline image was drawn to derive the per-patient total tumor data. Post-treatment image analysis was based on the same ROI as the corresponding baseline. Histogram metrics were extracted from both ROIs. Patients were categorized into the very good partial response or better (VGPR+) group and the less than VGPR group per the IMWG response criteria for response assessment. Progression-free survival (PFS) was also calculated. STATISTICAL TESTS: Mann-Whitney test and Fisher's exact or Chi-squared tests, Receiver operating characteristic (ROC) analysis and DeLong test, Kaplan-Meier analysis and Cox proportional hazards model. A two-tailed P-value <0.05 was considered statistically significant. RESULTS: Thirty patients were categorized into the VGPR+ group and six into the less than VGPR group. The ttADC histogram changes between post-treatment and baseline metrics (ΔttADC) revealed significant differences in all percentile values between the VGPR+ and less than VGPR groups. For distinguishing VGPR+, ΔttADC_5th percentile had the largest area under the curve (AUC) (0.950, 95% CI 0.821-0.995). Patients with lower ΔttADC_5th percentile values (cutoff point, 188.193) showed significantly longer PFS (HR = 34.911, 95% CI 6.392-190.677). DATA CONCLUSION: ttADC histogram may facilitate response assessment in patients with NDMM. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 4.

Autoimmune Pancreatitis Surveillance: A Simplified MRI Protocol Versus a Comprehensive Pancreatic MRI Protocol.

RATIONALE AND OBJECTIVES: Magnetic resonance imaging (MRI) has good diagnostic performance and causes no radiation damage, making it an ideal tool for the autoimmune pancreatitis (AIP) surveillance. However, its time cost is high. This study aimed to evaluate (1) whether a simplified protocol (SP) of MRI for AIP surveillance provides information equivalent to the comprehensive protocol (CP) and (2) the time cost reductions associated with using an SP. MATERIALS AND METHODS: This retrospective single-institutional study included 40 patients with AIP with at least two contrast-enhanced MRI/magnetic resonance cholangiopancreatography studies. Two radiologists evaluated two imaging sets (CP/SP) per patient, independently. Intra- and inter-observer agreement in the evaluation of the pancreas and extrapancreatic organs involvement using the SP/CP in addition to the time cost differences between the SP and CP were assessed. Intra- and inter-rater reliability were assessed using Cohen's kappa test, intraclass correlations, or the weighted kappa test. The differences in time costs between the CP and SP were compared using the Mann-Whitney U test or Wilcoxon signed-rank test. RESULTS: The SP had strong intra- and inter-observer agreement with the CP in evaluating MRI parameters (κ ï¼ž 0.60, moderate to excellent) and disease activity status (κ ï¼ž 0.80, all excellent). The overall image acquisition time cost for the SP was 49.2% of the CP. For the two radiologists, the image interpretation time cost of the SP was reduced by approximately 35% and 27% compared to the CP. CONCLUSION: For AIP surveillance, SP MRI provides information consistent with the CP and is less time-consuming.

Should All Pancreatic Cystic Lesions with Worrisome or High-Risk Features Be Resected? A Clinical and Radiological Machine Learning Model May Help to Answer.

RATIONALE AND OBJECTIVES: According to current guidelines, pancreatic cystic lesions (PCLs) with worrisome or high-risk features may have overtreatment. The purpose of this study was to build a clinical and radiological based machine-learning (ML) model to identify malignant PCLs for surgery among preoperative PCLs with worrisome or high-risk features. MATERIALS AND METHODS: Clinical and radiological details of 317 pathologically confirmed PCLs with worrisome or high-risk features were retrospectively analyzed and applied to ML models including Support Vector Machine, Logistic Regression (LR), Decision Tree, Bernoulli NB, Gaussian NB, K Nearest Neighbors and Linear Discriminant Analysis. The diagnostic ability for malignancy of the optimal model with the highest diagnostic AUC in the cross-validation procedure was further evaluated in internal (n = 77) and external (n = 50) testing cohorts, and was compared to two published guidelines in internal mucinous cyst cohort. RESULTS: Ten clinical and radiological feature-based LR model was the optimal model with the highest AUC (0.951) in the cross-validation procedure. In the internal testing cohort, LR model reached an AUC, accuracy, sensitivity, and specificity of 0.927, 0.909, 0.914, and 0.905; in the external testing cohort, LR model reached 0.948, 0.900, 0.963, and 0.826. When compared to the European guidelines and the ACG guidelines, LR model demonstrated significantly better accuracy and specificity in identifying malignancy, while maintaining the same high sensitivity. CONCLUSION: Clinical- and radiological-based LR model can accurately identify malignant PCLs in patients with worrisome or high-risk features, possessing diagnostic performance better than the European guidelines as well as ACG guidelines.

Prediction of Hepatic Encephalopathy After Transjugular Intrahepatic Portosystemic Shunt Based on CT Radiomic Features of Visceral Adipose Tissue.

RATIONALE AND OBJECTIVES: To evaluate the performance and clinical utility of CT radiomic features of visceral adipose tissue (VAT) in the prediction of hepatic encephalopathy (HE) after transjugular intrahepatic portosystemic shunt (TIPS). MATERIALS AND METHODS: This multi-center study was retrospectively designed. Patients with cirrhosis who underwent TIPS were recruited from January 2015 to December 2020. Pre-TIPS contrast-enhanced CT images were collected for VAT segmentation and radiomic feature extraction. Least absolute shrinkage and selection operator regression with ten-fold cross-validation was performed to reduce dimension. Logistic regression with regularization, support vector machine, and random forest were used for model construction. RESULTS: A total of 130 patients (90 men; mean age, 54 ± 11 years) were finally enrolled. The cohort was split into 85 patients for the training set (58 men; mean age, 53 ± 12 years) with 19 HE, 21 patients for the internal test set (17 men; mean age, 53 ± 11 years) with 5 HE, and 24 patients for the external test set (15 men; mean age, 55 ± 11 years). Ten radiomic features and C-reactive protein constituted radiomic-clinical models with the best performance. The average area under the receiver operating characteristic curve is 0.97 in the training set and 0.84 in the test sets. For a fixed sensitivity of 0.90, the specificity and negative predictive value of the model is 0.63 and 1.00, respectively; while for a fixed specificity of 0.90, the sensitivity and positive predictive value is 0.60 and 0.75, respectively. CONCLUSION: Machine learning models based on CT radiomic features extracted from VAT can predict post-TIPS HE with satisfactory performance. CLINICAL RELEVANCE STATEMENT: Our machine learning models based on CT radiomic features of visceral adipose tissue in patients with cirrhosis may assist in predicting hepatic encephalopathy after transjugular intrahepatic portosystemic shunt, indicating its potential in patient selection and clinical decision-making. KEY POINTS: Radiomics of visceral adipose tissue provide great help in predicting hepatic encephalopathy after transjugular intrahepatic portosystemic shunt. The clinical-radiomic models showed satisfactory performance with an average area under the receiver operating characteristic curve of 0.84. The model can hypothetically provide 90% sensitivity and 100% negative predictive value for guiding patients who are considering transjugular intrahepatic portosystemic shunt.

Simulation video: a tool to evaluate communications skills in radiologist residents.

BACKGROUND: Effective communication is a crucial component of radiology resident training, and many different aspects need to be explored when teaching and evaluating communication skills. To ensure that radiology residents' communication skill levels can be measured accurately, a standardized evaluation tool has been introduced. In twenty hospitals in Beijing, simulation videos have been developed as a way to assess the communication skills of radiology residents during their certification exams, to minimize evaluating biases. This study aims to assess the performance of a simulation video model in evaluating communications skills compared to the standard patient model. METHODS: This is a retrospective observational study. The performance of standard patient and simulation video models was evaluated through an eight-year examination of communication skills in radiology residents. From 2014 to 2021, communications skill tests were administered to 1003 radiology residents in 20 hospitals in Beijing. The standardized patient (SP) model was applied in 2014, and simulation videos were used from 2015 to 2021. The difficulty and discrimination radio of the tests were evaluated. The subjective survey for candidates on two models of communication skills evaluation was performed and analyzed. RESULTS: The simulation video model evaluation demonstrated stable difficulty (ranging from 0.92 to 0.98) and discrimination ratio (ranging from 0.37 to 0.49), except for minor exceptions of discrimination in 2019 (0.58) and 2020 (0.20). Furthermore, the Kruskal-Wallis H test revealed no significant differences in average scores between 2016 (93.9 ± 4.6) and 2018 (94.5 ± 4.2), 2016 and 2019 (97.3 ± 3.9), 2017 (97.0 ± 5.6) and 2019, 2017 and 2020 (97.7 ± 4.7), as well as 2019 and 2020 exams (all p ≥ 0.05). In addition, candidates who responded to the survey preferred the simulation video model (with a 77.2% response rate), with 62.7% choosing it over the SP model for communication skills evaluation. CONCLUSION: The simulation video demonstrated a stable and better acceptable construct for assessing radiology residents' communication skills.

A systematic review and meta-analysis of CT and MRI radiomics in ovarian cancer: methodological issues and clinical utility.

OBJECTIVES: We aimed to present the state of the art of CT- and MRI-based radiomics in the context of ovarian cancer (OC), with a focus on the methodological quality of these studies and the clinical utility of these proposed radiomics models. METHODS: Original articles investigating radiomics in OC published in PubMed, Embase, Web of Science, and the Cochrane Library between January 1, 2002, and January 6, 2023, were extracted. The methodological quality was evaluated using the radiomics quality score (RQS) and Quality Assessment of Diagnostic Accuracy Studies 2 (QUADAS-2). Pairwise correlation analyses were performed to compare the methodological quality, baseline information, and performance metrics. Additional meta-analyses of studies exploring differential diagnoses and prognostic prediction in patients with OC were performed separately. RESULTS: Fifty-seven studies encompassing 11,693 patients were included. The mean RQS was 30.7% (range - 4 to 22); less than 25% of studies had a high risk of bias and applicability concerns in each domain of QUADAS-2. A high RQS was significantly associated with a low QUADAS-2 risk and recent publication year. Significantly higher performance metrics were observed in studies examining differential diagnosis; 16 such studies as well as 13 exploring prognostic prediction were included in a separate meta-analysis, which revealed diagnostic odds ratios of 25.76 (95% confidence interval (CI) 13.50-49.13) and 12.55 (95% CI 8.38-18.77), respectively. CONCLUSION: Current evidence suggests that the methodological quality of OC-related radiomics studies is unsatisfactory. Radiomics analysis based on CT and MRI showed promising results in terms of differential diagnosis and prognostic prediction. CRITICAL RELEVANCE STATEMENT: Radiomics analysis has potential clinical utility; however, shortcomings persist in existing studies in terms of reproducibility. We suggest that future radiomics studies should be more standardized to better bridge the gap between concepts and clinical applications.

CT-based radiomics nomogram for prediction of survival after transarterial chemoembolization with drug-eluting beads in patients with hepatocellular carcinoma and portal vein tumor thrombus.

OBJECTIVES: To develop and validate a CT-based radiomics model for the prediction of the overall survival (OS) of patients with hepatocellular carcinoma (HCC) and portal vein tumor thrombus (PVTT) treated with drug-eluting beads transarterial chemoembolization (DEB-TACE). METHODS: Patients were retrospectively enrolled from two institutions for the constitution of training (n = 69) and validation (n = 31) cohorts with a median follow-up of 15 months. A total of 396 radiomics features were extracted from each baseline CT image. Features selected by variable importance and minimal depth were used for random survival forest model construction. The performance of the model was assessed using the concordance index (C-index), calibration curves, integrated discrimination index (IDI), net reclassification index (NRI), and decision curve analysis. RESULTS: Type of PVTT and tumor number were proved to be significant clinical indicators for OS. Arterial phase images were used to extract radiomics features. Three radiomics features were selected for model construction. The C-index for the radiomics model was 0.759 in the training cohort and 0.730 in the validation cohort. To improve the predictive performance, clinical indicators were integrated into the radiomics model to form a combined model with a C-index of 0.814 in the training cohort and 0.792 in the validation cohort. The IDI was significant in both cohorts for the combined model versus the radiomics model in predicting 12-month OS. CONCLUSIONS: Type of PVTT and tumor number affected the OS of HCC patients with PVTT treated with DEB-TACE. Moreover, the combined clinical-radiomics model had a satisfactory performance. CLINICAL RELEVANCE STATEMENT: A CT-based radiomics nomogram, which consisted of 3 radiomics features and 2 clinical indicators, was recommended to predict 12-month overall survival of patients with hepatocellular carcinoma and portal vein tumor thrombus initially treated with drug-eluting beads transarterial chemoembolization. KEY POINTS: • Type of portal vein tumor thrombus and tumor number were significant predictors of the OS. • Integrated discrimination index and net reclassification index provided a quantitative evaluation of the incremental impact added by new indicators for the radiomics model. • A nomogram based on a radiomics signature and clinical indicators showed satisfactory performance in predicting OS after DEB-TACE.

Seeing beyond the tumor: computed tomography image-based radiomic analysis helps identify ovarian clear cell carcinoma subtype in epithelial ovarian cancer.

OBJECTIVE: To develop and validate a model that can preoperatively identify the ovarian clear cell carcinoma (OCCC) subtype in epithelial ovarian cancer (EOC) using CT imaging radiomics and clinical data. MATERIAL AND METHODS: We retrospectively analyzed data from 282 patients with EOC (training set = 225, testing set = 57) who underwent pre-surgery CT examinations. Patients were categorized into OCCC or other EOC subtypes based on postoperative pathology. Seven clinical characteristics (age, cancer antigen [CA]-125, CA-199, endometriosis, venous thromboembolism, hypercalcemia, stage) were collected. Primary tumors were manually delineated on portal venous-phase images, and 1218 radiomic features were extracted. The F-test-based feature selection method and logistic regression algorithm were used to build the radiomic signature, clinical model, and integrated model. To explore the effects of integrated model-assisted diagnosis, five radiologists independently interpreted images in the testing set and reevaluated cases two weeks later with knowledge of the integrated model's output. The diagnostic performances of the predictive models, radiologists, and radiologists aided by the integrated model were evaluated. RESULTS: The integrated model containing the radiomic signature (constructed by four wavelet radiomic features) and three clinical characteristics (CA-125, endometriosis, and hypercalcinemia), showed better diagnostic performance (AUC = 0.863 [0.762-0.964]) than the clinical model (AUC = 0.792 [0.630-0.953], p = 0.295) and the radiomic signature alone (AUC = 0.781 [0.636-0.926], p = 0.185). The diagnostic sensitivities of the radiologists were significantly improved when using the integrated model (p = 0.023-0.041), while the specificities and accuracies were maintained (p = 0.074-1.000). CONCLUSION: Our integrated model shows great potential to facilitate the early identification of the OCCC subtype in EOC, which may enhance subtype-specific therapy and clinical management.

Extracellular volume fraction of the pancreas predicts glucose intolerance in patients undergoing major pancreatic surgeries.

PURPOSE: Pancreatic T1 value and extracellular volume fraction (ECV) are potential imaging biomarkers for pancreatic exocrine and endocrine function. This study aims to evaluate the ability of native T1 value and ECV of the pancreas in predicting postoperative new-onset diabetes (NODM) and worsened glucose tolerance in patients undergoing major pancreatic surgeries. METHODS: This retrospective study involved 73 patients who underwent 3 T pancreatic MRI with pre- and postcontrast T1 mapping before major pancreatic surgeries. Patients were divided into non-diabetic, pre-diabetic and diabetic groups based on their glycated hemoglobin (HbA1c) value. Preoperative native T1 value and ECV of the pancreas were compared among the three groups. The correlation of pancreatic T1 value and ECV with HbA1c was assessed by linear regression analysis, and the ability of pancreatic T1 value and ECV for predicting postoperative NODM and worsened glucose tolerance was assessed using Cox Proportional hazards regression analysis. RESULTS: Native pancreatic T1 value and ECV were both significantly higher in diabetic patients compared to pre-diabetic/non-diabetic patients, and ECV was also significantly higher in pre-diabetic patients compared to non-diabetic patients (all p < 0.05). Both native pancreatic T1 value and ECV showed positive correlation with preoperative HbA1c value (r = 0.50 and 0.55, respectively, both p < 0.001). ECV > 30.7% was the only independent predictor for NODM (HR = 5.687, 95% CI: 1.557, 13.468, p = 0.012) and worsened glucose tolerance (HR = 6.783, 95% CI:, 1.753, 15.842, p = 0.010) after surgery. CONCLUSIONS: Pancreatic ECV predicts the risk of postoperative NODM and worsened glucose tolerance in patients undergoing major pancreatic surgeries.

Transformer guided progressive fusion network for 3D pancreas and pancreatic mass segmentation.

Pancreatic masses are diverse in type, often making their clinical management challenging. This study aims to address the task of various types of pancreatic mass segmentation and detection while accurately segmenting the pancreas. Although convolution operation performs well at extracting local details, it experiences difficulty capturing global representations. To alleviate this limitation, we propose a transformer guided progressive fusion network (TGPFN) that utilizes the global representation captured by the transformer to supplement long-range dependencies lost by convolution operations at different resolutions. TGPFN is built on a branch-integrated network structure, where the convolutional neural network and transformer branches first perform separate feature extraction in the encoder, and then the local and global features are progressively fused in the decoder. To effectively integrate the information of the two branches, we design a transformer guidance flow to ensure feature consistency, and present a cross-network attention module to capture the channel dependencies. Extensive experiments with nnUNet (3D) show that TGPFN improves the mass segmentation (Dice: 73.93% vs. 69.40%) and detection accuracy (detection rate: 91.71% vs. 84.97%) on 416 private CTs, and also obtains performance improvements of mass segmentation (Dice: 43.86% vs. 42.07%) and detection (detection rate: 83.33% vs. 71.74%) on 419 public CTs.