Role of Additional MRI-Based Morphologic Measurements on the Performance of VI-RADS for Muscle-Invasive Bladder Cancer.

BACKGROUND: Vesical Imaging-Reporting and Data System (VI-RADS) is a pathway for the standardized imaging and reporting of bladder cancer staging using multiparametric (mp) MRI. PURPOSE: To investigate additional role of morphological (MOR) measurements to VI-RADS for the detection of muscle-invasive bladder cancer (MIBC) with mpMRI. STUDY TYPE: Retrospective. POPULATION: A total of 198 patients (72 MIBC and 126 NMIBC) underwent bladder mpMRI was included. FIELD STRENGTH/SEQUENCE: 3.0 T/T2-weighted imaging with fast-spin-echo sequence, spin-echo-planar diffusion-weighted imaging and dynamic contrast-enhanced imaging with fast 3D gradient-echo sequence. ASSESSMENT: VI-RADS score and MOR measurement including tumor location, number, stalk, cauliflower-like surface, type of tumor growth, tumor-muscle contact margin (TCM), tumor-longitudinal length (TLL), and tumor cellularity index (TCI) were analyzed by three uroradiologists (3-year, 8-year, and 15-year experience of bladder MRI, respectively) who were blinded to histopathology. STATISTICAL TESTS: Significant MOR measurements associated with MIBC were tested by univariable and multivariable logistic regression (LR) analysis with odds ratio (OR). Area under receiver operating characteristic curve (AUC) with DeLong's test and decision curve analysis (DCA) were used to compared the performance of unadjusted vs. adjusted VI-RADS. A P-value <0.05 was considered statistically significant. RESULTS: TCM (OR 9.98; 95% confidence interval [CI] 4.77-20.8), TCI (OR 5.72; 95% CI 2.37-13.8), and TLL (OR 3.35; 95% CI 1.40-8.03) were independently associated with MIBC at multivariable LR analysis. VI-RADS adjusted by three MORs achieved significantly higher AUC (reader 1 0.908 vs. 0.798; reader 2 0.906 vs. 0.855; reader 3 0.907 vs. 0.831) and better clinical benefits than unadjusted VI-RADS at DCA. Specially in VI-RADS-defined equivocal lesions, MOR-based adjustment resulted in 55.5% (25/45), 70.4% (38/54), and 46.4% (26/56) improvement in accuracy for discriminating MIBC in three readers, respectively. DATA CONCLUSION: MOR measurements improved the performance of VI-RADS in detecting MIBC with mpMRI, especially for equivocal lesions. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY: Stage 2.

Biopsy-free AI-aided precision MRI assessment in prediction of prostate cancer biochemical recurrence.

BACKGROUND: To investigate the predictive ability of high-throughput MRI with deep survival networks for biochemical recurrence (BCR) of prostate cancer (PCa) after prostatectomy. METHODS: Clinical-MRI and histopathologic data of 579 (train/test, 463/116) PCa patients were retrospectively collected. The deep survival network (iBCR-Net) is based on stepwise processing operations, which first built an MRI radiomics signature (RadS) for BCR, and predicted the T3 stage and lymph node metastasis (LN+) of tumour using two predefined AI models. Subsequently, clinical, imaging and histopathological variables were integrated into iBCR-Net for BCR prediction. RESULTS: RadS, derived from 2554 MRI features, was identified as an independent predictor of BCR. Two predefined AI models achieved an accuracy of 82.6% and 78.4% in staging T3 and LN+. The iBCR-Net, when expressed as a presurgical model by integrating RadS, AI-diagnosed T3 stage and PSA, can match a state-of-the-art histopathological model (C-index, 0.81 to 0.83 vs 0.79 to 0.81, p > 0.05); and has maximally 5.16-fold, 12.8-fold, and 2.09-fold (p < 0.05) benefit to conventional D'Amico score, the Cancer of the Prostate Risk Assessment (CAPRA) score and the CAPRA Postsurgical score. CONCLUSIONS: AI-aided iBCR-Net using high-throughput MRI can predict PCa BCR accurately and thus may provide an alternative to the conventional method for PCa risk stratification.

Diagnosis, treatment and genetic analysis of an elderly patient with Gaucher's disease characterized by marked multiple bone destruction.

Gaucher's disease is a rare autosomal recessive genetic disease. Due to the decrease or lack of glucocerebrosidase (GBA) activity in lysosome caused by the mutation of GBA gene, its substrate glucocerebroside is detained in lysosome, resulting in clinical manifestations of liver, spleen, kidney, bone, hematopoietic system and even nervous system involvement. Here, we report a case of elderly patient presenting marked multiple bone destruction, with childhood medical history of splenectomy and "osteomyelitis". The patient has a significantly enlarged liver, accompanied by anemia, thrombocytopenia and osteopenia. Laboratory studies show this patient has low blood GBA activity and high glucosyl sphingosine level and increased chitotriosidase activity. Genetic testing revealed a homozygous missense variant NM_001005741.2 c.770A>G (p.Asp257Gly) in the patient's GBA gene. After 6 months of enzyme replacement therapy, the patient's platelets returned to normal, anemia improved, and liver volume decreased. Further detections show that the mother and brothers of the patient have heterozygous mutations at this locus, which is consistent with Mendelian inheritance law. Although this variant has not been reported in literatures or database, both clinical manifestations, characteristics of enzymology and biomarkers, and the effect of enzyme replacement therapy support the diagnosis of Gaucher's disease. The Asp257Gly variant is therefore assessed as a clinical pathogenic variant. This study expands the spectrum of the GBA gene variants. The diagnosis and treatment process of this case also provide reference for the early identification, diagnosis and early treatment of this kind of patients.

Improving the understanding of PI-RADS in practice: characters of PI-RADS 4 and 5 lesions with negative biopsy.

The Prostate Imaging Reporting and Data System (PI-RADS) has good ability to identify the nature of lesions on prostate magnetic resonance imaging (MRI). However, some lesions are still reported as PI-RADS 4 and 5 but are biopsy-proven benign. Herein, we aimed to summarize the reasons for the negative prostate biopsy of patients who were assessed as PI-RADS 4 and 5 by biparameter MRI. We retrospectively sorted out the prostate MRI, treatment, and follow-up results of patients who underwent a biparameter MRI examination of the prostate in The First Affiliated Hospital of Nanjing Medical University (Nanjing, China) from August 2019 to June 2021 with PI-RADS 4 and 5 but a negative biopsy. We focused on reviewing the MRI characteristics. A total of 467 patients underwent transperineal prostate biopsy. Among them, biopsy pathology of 93 cases were negative. After follow-up, 90 patients were ruled out of prostate cancer. Among the 90 cases, 40 were considered to be overestimated PI-RADS after review. A total of 22 cases were transition zone (TZ) lesions with regular appearance and clear boundaries, and 3 cases were symmetrical lesions. Among 15 cases, the TZ nodules penetrated the peripheral zone (PZ) and were mistaken for the origin of PZ. A total of 17 cases of lesions were difficult to distinguish from prostate cancer. Among them, 5 cases were granulomatous inflammation (1 case of prostate tuberculosis). A total of 33 cases were ambiguous lesions, whose performance was between PI-RADS 3 and 4. In summary, the reasons for "false-positive MRI diagnosis" included PI-RADS overestimation, ambiguous images giving higher PI-RADS, diseases that were really difficult to distinguish, and missed lesion in the initial biopsy; and the first two accounted for the most.

LiSNet: An artificial intelligence -based tool for liver imaging staging of hepatocellular carcinoma aggressiveness.

BACKGROUND: Presurgical assessment of hepatocellular carcinoma (HCC) aggressiveness can benefit patients' treatment options and prognosis. PURPOSE: To develop an artificial intelligence (AI) tool, namely, LiSNet, in the task of scoring and interpreting HCC aggressiveness with computed tomography (CT) imaging. METHODS: A total of 358 patients with HCC undergoing curative liver resection were retrospectively included. Three subspecialists were recruited to pixel-wise annotate and grade tumor aggressiveness based on CT imaging. LiSNet was trained and validated in 193 and 61 patients with a deep neural network to emulate the diagnostic acumen of subspecialists for staging HCC. The test set comprised 104 independent patients. We subsequently compared LiSNet with an experience-based binary diagnosis scheme and human-AI partnership that combined binary diagnosis and LiSNet for assessing tumor aggressiveness. We also assessed the efficiency of LiSNet for predicting survival outcomes. RESULTS: At the pixel-wise level, the agreement rate of LiSNet with subspecialists was 0.658 (95% confidence interval [CI]: 0.490-0.779), 0.595 (95% CI: 0.406-0.734), and 0.369 (95% CI: 0.134-0.566), for scoring HCC aggressiveness grades I, II, and III, respectively. Additionally, LiSNet was comparable to subspecialists for predicting histopathological microvascular invasion (area under the curve: LiSNet: 0.668 [95% CI: 0.559-0.776] versus subspecialists: 0.699 [95% CI: 0.591-0.806], p > 0.05). In a human-AI partnered diagnosis, combining LiSNet and experience-based binary diagnosis can achieve the best predictive ability for microvascular invasion (area under the curve: 0.705 [95% CI: 0.589-0.820]). Furthermore, LiSNet was able to indicate overall survival after surgery. CONCLUSION: The designed LiSNet tool warrants evaluation as an alternative tool for radiologists to conduct automatic staging of HCC aggressiveness at the pixel-wise level with CT imaging. Its prognostic value might benefit patients' treatment options and survival prediction.

Non-uniform model of relationship between surface strain and rust expansion force of reinforced concrete.

When operating within the environments rich with sodium chloride, steel bars of reinforced concrete structures are often subject to corrosion caused by surrounding erosive materials, and the associated rust expansion force due to corrosion takes a critical role in determining the durability of relevant reinforced concrete structures. By investigating the corrosion course of steel reinforcement with theory of elasticity, a numerical rust expansion model is established for the moment of concrete surface rupture based on non-uniform sin function. Cuboid reinforced concrete specimen with squared cross sections is tested to analyze the rust expansion when concrete cracks due to corrosive forces. The utility of the established expansion model is validated by numerical simulation with Abaqus through the comparison between the associated outcomes. The impacts of steel bar diameter and concrete cover thickness on the magnitude of rust expansion force are discussed.

Prediction of prognostic risk factors in hepatocellular carcinoma with transarterial chemoembolization using multi-modal multi-task deep learning.

BACKGROUND: Due to heterogeneity of hepatocellular carcinoma (HCC), outcome assessment of HCC with transarterial chemoembolization (TACE) is challenging. METHODS: We built histologic-related scores to determine microvascular invasion (MVI) and Edmondson-Steiner grade by training CT radiomics features using machine learning classifiers in a cohort of 494 HCCs with hepatic resection. Meanwhile, we developed a deep learning (DL)-score for disease-specific survival by training CT imaging using DL networks in a cohort of 243 HCCs with TACE. Then, three newly built imaging hallmarks with clinicoradiologic factors were analyzed with a Cox-Proportional Hazard (Cox-PH) model. FINDINGS: In HCCs with hepatic resection, two imaging hallmarks resulted in areas under the curve (AUCs) of 0.79 (95% confidence interval [CI]: 0.71-0.85) and 0.72 (95% CI: 0.64-0.79) for predicting MVI and Edmondson-Steiner grade, respectively, using test data. In HCCs with TACE, higher DL-score (hazard ratio [HR]: 3.01; 95% CI: 2.02-4.50), American Joint Committee on Cancer (AJCC) stage III+IV (HR: 1.71; 95% CI: 1.12-2.61), Response Evaluation Criteria in Solid Tumors (RECIST) with stable disease + progressive disease (HR: 2.72; 95% CI: 1.84-4.01), and TACE-course > 3 (HR: 0.65; 95% CI: 0.45-0.76) were independent prognostic factors. Using these factors via a Cox-PH model resulted in a concordance index of 0.73 (95% CI: 0.71-0.76) for predicting overall survival and AUCs of 0.85 (95% CI: 0.81-0.89), 0.90 (95% CI: 0.86-0.94), and 0.89 (95% CI: 0.84-0.92), respectively, for predicting 3-year, 5-year, and 10-year survival. INTERPRETATION: Our study offers a DL-based, noninvasive imaging hallmark to predict outcome of HCCs with TACE. FUNDING: This work was supported by the key research and development program of Jiangsu Province (Grant number: BE2017756).

Direct Comparison of Four Presurgical Stratifying Schemes for Prediction of Microvascular Invasion in Hepatocellular Carcinoma by Gadoxetic Acid-Enhanced MRI.

BACKGROUND: Microvascular invasion (MVI) is implicated in the poor prognosis of hepatocellular carcinoma (HCC). Presurgical stratifying schemes have been proposed for HCC-MVI but lack external validation. PURPOSE: To perform external validation and comparison of four presurgical stratifying schemes for the prediction of MVI using gadoxetic acid-based MRI in a cohort of HCC patients. STUDY TYPE: Retrospective. SUBJECTS: Included were 183 surgically resected HCCs from patients who underwent pretreatment MRI. FIELD STRENGTH/SEQUENCE: This includes 1.5-3.0 T with T2 , T1 , diffusion-weighted imaging (DWI), and dynamic gadoxetic acid contrast-enhancement imaging sequences. ASSESSMENT: A two-trait predictor of venous invasion (TTPVI), Lei model, Lee model, and Xu model were compared. We relied on preoperative characteristics and imaging findings via four independent radiologists who were blinded to histologic results, as required by the tested tools. STATISTICAL TEST: Tests of accuracy between predicted and observed HCC-MVI rates using receiver operating characteristic (ROC) curve and decision curve analysis. The intraclass correlation coefficient (ICC) and Cronbach's alpha statistics were used to evaluate reproducibility. RESULTS: HCC-MVI was identified in 52 patients (28.4%). The average ROC curves (AUCs) for HCC-MVI predictions were 0.709-0.880, 0.714-0.828, and 0.588-0.750 for the Xu model, Lei model, and Lee model, respectively. The rates of accuracy were 60.7-81.4%, 69.9-75.9%, and 65.6-73.8%, respectively. Decision curve analyses indicated a higher benefit for the Xu and Lei models compared to the Lee model. The ICC and Cronbach's alpha index were highest in the Lei model (0.896/0.943), followed by the Xu model (0.882/0.804), and the Lee model (0.769/0.715). The TTPVI resulted in a Cronbach's alpha index of 0.606 with a sensitivity of 34.6-61.5% and a specificity of 76.3-91.6%. DATA CONCLUSION: Stratifying schemes relying on gadoxetic acid-enhanced MRI provide an additional insight into the presence of preoperative MVI. The Xu model outperformed the other models in terms of accuracy when performed by an experienced radiologist. Conversely, the Lei model outperformed the other models in terms of reproducibility. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2 J. Magn. Reson. Imaging 2020;52:433-447.

Radiomic Features at Contrast-enhanced CT Predict Recurrence in Early Stage Hepatocellular Carcinoma: A Multi-Institutional Study.

Background Early stage hepatocellular carcinoma (HCC) is the ideal candidate for resection in patients with preserved liver function; however, cancer will recur in half of these patients and no reliable prognostic tool has been established. Purpose To investigate the effectiveness of radiomic features in predicting tumor recurrence after resection of early stage HCC. Materials and Methods In total, 295 patients (median age, 58 years; interquartile range, 50-65 years; 221 men) who underwent contrast material-enhanced CT and curative resection for early stage HCC that met the Milan criteria between February 2009 and December 2016 were retrospectively recruited from three independent institutions. Follow-up consisted of serum α-fetoprotein level, liver function tests, and dynamic imaging examinations every 3 months during the first 2 years and then every 6 months thereafter. In the development cohort of 177 patients from institution 1, recurrence-related radiomic features were computationally extracted from the tumor and its periphery and a radiomics signature was built with least absolute shrinkage and selection operator regression. Two models, one integrating preoperative and one integrating pre- and postoperative variables, were created by using multivariable Cox regression analysis. An independent external cohort of 118 patients from institutions 2 and 3 was used to validate the proposed models. Results The preoperative model integrated radiomics signature with serum α-fetoprotein level and tumor number; the postoperative model incorporated microvascular invasion and satellite nodules into the above-mentioned predictors. In both study cohorts, two radiomics-based models provided better predictive performance (concordance index ≥0.77, P < .05 for all), lower prediction error (integrated Brier score ≤0.14), and larger net benefits, as determined by means of decision curve analysis, than rival models without radiomics and widely adopted staging systems. The radiomics-based models gave three risk strata with high, intermediate, or low risk of recurrence and distinct profiles of recurrent tumor number. Conclusion The proposed radiomics models with pre- and postresection features helped predict tumor recurrence for early stage hepatocellular carcinoma. © RSNA, 2020 Online supplemental material is available for this article.

Radiologic-Radiomic Machine Learning Models for Differentiation of Benign and Malignant Solid Renal Masses: Comparison With Expert-Level Radiologists.

OBJECTIVE. The objective of our study was to compare the performance of radiologicradiomic machine learning (ML) models and expert-level radiologists for differentiation of benign and malignant solid renal masses using contrast-enhanced CT examinations. MATERIALS AND METHODS. This retrospective study included a cohort of 254 renal cell carcinomas (RCCs) (190 clear cell RCCs [ccRCCs], 38 chromophobe RCCs [chrRCCs], and 26 papillary RCCs [pRCCs]), 26 fat-poor angioleiomyolipomas, and 10 oncocytomas with preoperative CT examinations. Lesions identified by four expert-level radiologists (> 3000 genitourinary CT and MRI studies) were manually segmented for radiologicradiomic analysis. Disease-specific support vector machine radiologic-radiomic ML models for classification of renal masses were trained and validated using a 10-fold cross-validation. Performance values for the expert-level radiologists and radiologic-radiomic ML models were compared using the McNemar test. RESULTS. The performance values for the four radiologists were as follows: sensitivity of 73.7-96.8% (median, 84.5%; variance, 122.7%) and specificity of 48.4-71.9% (median, 61.8%; variance, 161.6%) for differentiating ccRCCs from pRCCs and chrRCCs; sensitivity of 73.7-96.8% (median, 84.5%; variance, 122.7%) and specificity of 52.8-88.9% for differentiating ccRCCs from fat-poor angioleiomyolipomas and oncocytomas (median, 80.6%; variance, 269.1%); and sensitivity of 28.1-60.9% (median, 84.5%; variance, 122.7%) and specificity of 75.0-88.9% for differentiating pRCCs and chrRCCs from fat-poor angioleiomyolipomas and oncocytomas (median, 50.0%; variance, 191.1%). After a 10-fold cross-validation, the radiologic-radiomic ML model yielded the following performance values for differentiating ccRCCs from pRCCs and chrRCCs, ccRCCs from fat-poor angioleiomyolipomas and oncocytomas, and pRCCs and chrRCCs from fat-poor angioleiomyolipomas and oncocytomas: a sensitivity of 90.0%, 86.3%, and 73.4% and a specificity of 89.1%, 83.3%, and 91.7%, respectively. CONCLUSION. Expert-level radiologists had obviously large variances in performance for differentiating benign from malignant solid renal masses. Radiologic-radiomic ML can be a potential way to improve interreader concordance and performance.

Machine-learning analysis of contrast-enhanced CT radiomics predicts recurrence of hepatocellular carcinoma after resection: A multi-institutional study.

BACKGROUND: Current guidelines recommend surgical resection as the first-line option for patients with solitary hepatocellular carcinoma (HCC); unfortunately, postoperative recurrence rate remains high and there is no reliable prediction tool. We explored the potential of radiomics coupled with machine-learning algorithms to improve the predictive accuracy for HCC recurrence. METHODS: A total of 470 patients who underwent contrast-enhanced CT and curative resection for solitary HCC were recruited from 3 independent institutions. In the training phase of 210 patients from Institution 1, a radiomics-derived signature was generated based on 3384 engineered features extracted from primary tumor and its periphery using aggregated machine-learning framework. We employed Cox modeling to build predictive models. The models were then validated using an internal dataset of 107 patients and an external dataset of 153 patients from Institution 2 and 3. FINDINGS: Using the machine-learning framework, we identified a three-feature signature that demonstrated favorable prediction of HCC recurrence across all datasets, with C-index of 0.633-0.699. Serum alpha-fetoprotein, albumin-bilirubin grade, liver cirrhosis, tumor margin, and radiomics signature were selected for preoperative model; postoperative model incorporated satellite nodules into above-mentioned predictors. The two models showed superior prognostic performance, with C-index of 0.733-0.801 and integrated Brier score of 0.147-0.165, compared with rival models without radiomics and widely used staging systems (all P < 0.05); they also gave three risk strata for recurrence with distinct recurrence patterns. INTERPRETATION: When integrated with clinical data sources, our three-feature radiomics signature promises to accurately predict individual recurrence risk that may facilitate personalized HCC management.

A radiomics approach to predict lymph node metastasis and clinical outcome of intrahepatic cholangiocarcinoma.

OBJECTIVES: This study was conducted in order to establish and validate a radiomics model for predicting lymph node (LN) metastasis of intrahepatic cholangiocarcinoma (IHC) and to determine its prognostic value. METHODS: For this retrospective study, a radiomics model was developed in a primary cohort of 103 IHC patients who underwent curative-intent resection and lymphadenectomy. Radiomics features were extracted from arterial phase computed tomography (CT) scans. A radiomics signature was built based on highly reproducible features using the least absolute shrinkage and selection operator (LASSO) method. Multivariate logistic regression analysis was adopted to establish a radiomics model incorporating radiomics signature and other independent predictors. Model performance was determined by its discrimination, calibration, and clinical usefulness. The model was internally validated in 52 consecutive patients. RESULTS: The radiomics signature comprised eight LN-status-related features and showed significant association with LN metastasis in both cohorts (p < 0.001). A radiomics nomogram that incorporates radiomics signature and CA 19-9 level showed good calibration and discrimination in the primary cohort (AUC 0.8462) and validation cohort (AUC 0.8921). Promisingly, the radiomics nomogram yielded an AUC of 0.9224 in the CT-reported LN-negative subgroup. Decision curve analysis confirmed the clinical utility of this nomogram. High risk for metastasis portended significantly lower overall and recurrence-free survival than low risk for metastasis (both p < 0.001). The radiomics nomogram was an independent preoperative predictor of overall and recurrence-free survival. CONCLUSIONS: Our radiomics model provided a robust diagnostic tool for prediction of LN metastasis, especially in CT-reported LN-negative IHC patients, that may facilitate clinical decision-making. KEY POINTS: • The radiomics nomogram showed good performance for prediction of LN metastasis in IHC patients, particularly in the CT-reported LN-negative subgroup. • Prognosis of high-risk patients remains dismal after curative-intent resection. • The radiomics model may facilitate clinical decision-making and define patient subsets benefiting most from surgery.

Radiomic analysis of contrast-enhanced CT predicts microvascular invasion and outcome in hepatocellular carcinoma.

BACKGROUND & AIMS: Microvascular invasion (MVI) impairs surgical outcomes in patients with hepatocellular carcinoma (HCC). As there is no single highly reliable factor to preoperatively predict MVI, we developed a computational approach integrating large-scale clinical and imaging modalities, especially radiomic features from contrast-enhanced CT, to predict MVI and clinical outcomes in patients with HCC. METHODS: In total, 495 surgically resected patients were retrospectively included. MVI-related radiomic scores (R-scores) were built from 7,260 radiomic features in 6 target volumes. Six R-scores, 15 clinical factors, and 12 radiographic scores were integrated into a predictive model, the radiographic-radiomic (RR) model, with multivariate logistic regression. RESULTS: Radiomics related to tumor size and intratumoral heterogeneity were the top-ranked MVI predicting features. The related R-scores showed significant differences according to MVI status (p <0.001). Regression analysis identified 8 MVI risk factors, including 5 radiographic features and an R-score. The R-score (odds ratio [OR] 2.34) was less important than tumor capsule (OR 5.12), tumor margin (OR4.20), and peritumoral enhancement (OR 3.03). The RR model using these predictors achieved an area under the curve (AUC) of 0.909 in training/validation and 0.889 in the test set. Progression-free survival (PFS) and overall survival (OS) were significantly different between the RR-predicted MVI-absent and MVI-present groups (median PFS: 49.5 vs. 12.9 months; median OS: 76.3 vs. 47.3 months). RR-computed MVI probability, histologic MVI, tumor size, and Edmondson-Steiner grade were independently associated with disease-specific recurrence and mortality. CONCLUSIONS: The computational approach, integrating large-scale clinico-radiologic and radiomic features, demonstrates good performance for predicting MVI and clinical outcomes. However, radiomics with current CT imaging analysis protocols do not provide statistically significant added value to radiographic scores. LAY SUMMARY: The most effective treatment for hepatocellular carcinoma (HCC) is surgical removal of the tumor but often recurrence occurs, partly due to the presence of microvascular invasion (MVI). Lacking a single highly reliable factor able to preoperatively predict MVI, we developed a computational approach to predict MVI and the long-term clinical outcome of patients with HCC. In particular, the added value of radiomics, a newly emerging form of radiography, was comprehensively investigated. This computational method can enhance the communication with the patient about the likely success of the treatment and guide clinical management, with the aim of finding drugs that reduce the risk of recurrence.

Biliary Tract Cancer at CT: A Radiomics-based Model to Predict Lymph Node Metastasis and Survival Outcomes.

Purpose To evaluate a radiomics model for predicting lymph node (LN) metastasis in biliary tract cancers (BTCs) and to determine its prognostic value for disease-specific and recurrence-free survival. Materials and Methods For this retrospective study, a radiomics model was developed on the basis of a primary cohort of 177 patients with BTC who underwent resection and LN dissection between June 2010 and December 2016. Radiomic features were extracted from portal venous CT scans. A radiomics signature was built on the basis of reproducible features by using the least absolute shrinkage and selection operator method. Multivariable logistic regression model was adopted to establish a radiomics nomogram. Nomogram performance was determined by its discrimination, calibration, and clinical usefulness. The model was internally validated in 70 consecutive patients with BTC between January 2017 and February 2018. Results The radiomics signature, composed of three LN-status-related features, was associated with LN metastasis in primary and validation cohorts (P < .001). The radiomics nomogram that incorporated radiomics signature and CT-reported LN status showed good calibration and discrimination in primary cohort (area under the curve, 0.81) and validation cohort (area under the curve, 0.80). Patients at high risk of LN metastasis portended lower disease-specific and recurrence-free survival than did those at low risk after surgery (both P < .001). High-risk LN metastasis was an independent preoperative predictor of disease-specific survival (hazard ratio, 3.37; P < .001) and recurrence-free survival (hazard ratio, 1.98; P = .003). Conclusion A radiomics model derived from portal phase CT of the liver has good performance for predicting lymph node metastasis in biliary tract cancer and may help to improve clinical decision making. © RSNA, 2018 Online supplemental material is available for this article. See also the editorial by Laghi and Voena in this issue.

Radiologic evaluation of vasculobiliary anatomy in the umbilical fissure.

BACKGROUND: Preoperative evaluation of vasculobiliary anatomy in the umbilical fissure (U-point) is pivotal for perihilar cholangiocarcinoma (PCCA) applied to right-sided hepatectomy. The purpose of our study was to review the vasculobiliary anatomy in the U-point using three-dimensional (3D) reconstruction technique, to investigate the diagnostic ability of 2D scans to evaluate anatomic variations, and to discuss its surgical implications. METHODS: A retrospective study of 159 patients with Bismuth type I, II, and IIIa PCCA, who received surgery at our institution from November 2012 to September 2016, was conducted. Anatomic structures were assessed using multidetector computed tomography (MDCT) by one hepatobiliary surgeon, whereas 3D images were reconstructed by an independent radiologist. Normal confluence pattern of left biliary system was defined as the left medial segmental bile duct (B4) joining the common trunk of segment II (B2) and segment III (B3) ducts, whereas aberrant confluence patterns were classified into 3 types: type I, triple confluence of B2, B3, and B4; type II, B2 draining into the common trunk of B3 and B4; type III, other patterns. Surgical anatomy of B4 was classified into the central, peripheral, and combined type according to its relation to the hepatic confluence. The lengths from the bile duct branch of Spiegel's lobe (B1l) to the orifice of B4 and the junction of B2 and B3 were measured on 3D images. The anatomy of left hepatic artery (LHA) was classified according to different origins and the spatial relationship related to the U-point. RESULTS: 3D reconstruction revealed that normal confluence pattern of left biliary system was observed in 71.1% (113/159) of all patients, and variant patterns were type I in 11.9% (19/159), type II in 12.6% (20/159), and type III in 4.4% (7/159). The length from B1l to the junction of B2 and B3 was 12.1 ± 3.1 mm in type I variation, which was significantly shorter than that in normal configuration (30.0 ± 6.8 mm, P < 0.001) but significantly longer than that in type II variation (9.6 ± 3.4 mm, P = 0.019). Surgical anatomy of B4: the peripheral type was most commonly seen (74.2%, 118/159), followed by central type (15.7%, 25/159) and combined type (10.1%, 16/159). The distance between the B1l and B4 was 8.4 ± 2.4 mm in central and combined type, which was significantly shorter than that in peripheral type (14.5 ± 4.1 mm, P < 0.001). A replaced or accessory LHA from the left gastric artery was present in 6 (3.8%) and 9 (5.7%) patients, respectively. LHA running along the left caudal position of U-point was present in 143 cases (89.9%), along the right cranial position of U-point in nine cases (5.7 %), and combined position in seven cases (4.4%). Interobserver agreement of two imaging modalities was almost perfect in biliary confluence pattern (kappa = 0.90; 95% confidence interval: 0.79-1.00), substantial in surgical anatomy of B4 (kappa = 0.74; 95% confidence interval: 0.62-0.86), and perfect in LHA (kappa = 1.00). CONCLUSIONS: Thoroughly understanding the imaging characters of surgical anatomy in the U-point may be benefit for preoperative evaluation of PCCA by successive review of 2D images alone, whereas 3D reconstruction technique allows detailed hepatic anatomy and individualized surgical planning for advanced cases.

Clinical Implications of Biliary Confluence Pattern for Bismuth-Corlette Type IV Hilar Cholangiocarcinoma Applied to Hemihepatectomy.

BACKGROUND: Since biliary variations are commonly seen, our aims are to clarify these insidious variations and discuss their surgicopathologic implications for Bismuth-Corlette (BC) type IV hilar cholangiocarcinoma (HC) applied to hemihepatectomy. METHODS: Three-dimensional images of patients with distal bile duct obstruction (n = 97) and advanced HC (n = 79) were reconstructed and analyzed retrospectively. Normal biliary confluence pattern was defined as the peripheral segment IV duct (B4) joining the common trunk of segment II (B2) and segment III (B3) ducts to form the left hepatic duct (LHD) that then joined the right hepatic duct (RHD). The lengths from left and right secondary biliary ramifications to the right side of the umbilical portion of the left portal vein (Rl-L) and the cranio-ventral side of the right portal vein (Rr-R) were measured, respectively, and compared with the resectable bile duct length in HCs. Surgicopathologic findings were compared between different BC types. RESULTS: The resectable bile duct length in right hemihepatectomy for eradication of type IV tumors was significantly longer than the Rl-L length in normal biliary configuration (17.4 ± 1.8 and 10.3 ± 3.4 mm, respectively, p < 0.001), and type III variation (B2 joining the common trunk of B3 and B4) was the predominant configuration (53.8%). The resectable length in left hemihepatectomy for eradication of type IV tumors was comparable with the Rr-R length in RHD absent cases (15.2 ± 2.5 and 16.4 ± 2.6 mm, respectively, p = 0.177) but significantly longer than that in normal configuration (p < 0.001). The estimated length was 8.5 ± 2.0 mm in unresectable cases. There was no significant difference between type III and IV tumors, except for the rate of nodal metastasis (29.7 and 76.0%, respectively, p < 0.001). CONCLUSION: Hemihepatectomy might be selected for curative-intent resection of BC type IV tumors considering the advantageous biliary variations, whereas anatomical trisegmentectomy is recommended for the resectable bile duct length less than 10 mm. Biliary variations might result in excessive classification of BC type IV but require validation on further study.

Liver Imaging Reporting and Data System:: Substantial Discordance Between CT and MR for Imaging Classification of Hepatic Nodules.

RATIONALE AND OBJECTIVES: The Liver Imaging Reporting and Data System (LI-RADS) is a newly developed nomogram for standardizing the performance and interpretation of liver imaging. However, it is unclear which imaging technique is optimal to exactly define LI-RADS scale. This study aims to determine the concordance of computed tomography (CT) and magnetic resonance imaging (MRI) for the classification of hepatic nodules (HNs) using a LI-RADS scoring system. MATERIALS AND METHODS: Major imaging features (arterial hyper-enhancement, washout, pseudo-capsule, diameter, and tumor embolus) on CT versus MRI for 118 HNs in 84 patients with diffuse liver disease were rated independently using LI-RADS by two groups of readers. Inter-reader agreement (IRA) and intraclass agreement was determined by Fleiss and Cohen's kappa (κ). Logistic regression for correlated data was used to compare diagnostic ability. RESULTS: IRA was perfect for determination of nodule size and tumor embolus (κ = 0.94-0.98). IRA was moderate to substantial for determination of arterial hyper-enhancement, washout, and pseudo-capsule (κ = 0.54-0.72). Intraclass agreement between CT and MRI was substantial for determination of washout (0.632 [95% CI: 0.494, 0.771]) and pseudo-capsule (0.670 [95% CI: 0.494, 0.847]), and fair for arterial hyper-enhancement (0.203 [95% CI: 0.051, 0.354]). CT against MR produced false-negative findings of arterial hyper-enhancement by 57.1%, washout by 21.2%, and pseudo-capsule by 42.9%; and underestimated LI-RADS score by 16.9% for LR 3, 37.3% for LR 4, and 8.5% for LR 5. CT produced significantly lower accuracy (54.3% vs 67.8%, P < 0.001) and sensitivity (31.6% vs 71.1%, P < 0.001) than MRI in the prediction of malignancy. CONCLUSIONS: There are substantial discordance between CT and MR for stratification of HNs using LI-RADS. MRI could be better than CT in optimizing the performance of LI-RADS.

Classifying CT/MR findings in patients with suspicion of hepatocellular carcinoma: Comparison of liver imaging reporting and data system and criteria-free Likert scale reporting models.

PURPOSE: To compare the Liver Imaging Reporting and Data System (LI-RADS) and a criteria-free Likert scale (LS) reporting models for classifying computed tomography/magnetic resonance imaging (CT/MR) findings of suspicious hepatocellular carcinoma (HCC). MATERIALS AND METHODS: Imaging data of 281 hepatic nodules in 203 patients were retrospectively included. Imaging characteristics including diameter, arterial hyperenhancement, washout, and capsule were reviewed independently by two groups of readers using LI-RADS and LS (range, score 1-5). LS is primarily based on the overall impression of image findings without using fixed criteria. Interreader agreement (IRA), intraclass agreement (ICA), and diagnostic performance were determined by Fleiss, Cohen's kappa (κ), and logistic regression, respectively. RESULTS: There were 167 contrast-enhanced CT (CECT) versus 114 MR data. Overall, IRA was moderate (κ = 0.47, 0.52); IRA was moderate-to-good for arterial hyperenhancement, washout, and capsule (κ = 0.56-0.69); excellent for diameter and tumor embolus (κ = 0.99). Overall, ICA between LI-RADS and LS was moderate (κ = 0.44-0.50); ICA was good for scores 1-2 (κ = 0.71-0.90), moderate for scores 3 and 5 (κ = 0.41-0.52), but very poor for score 4 (κ = 0.11-0.19). LI-RADS produced significantly lower accuracy (78.6% vs. 87.2%) and sensitivity (72.1% vs. 92.8%), higher specificity (97.3% vs. 71.2%) and positive likelihood ratio (+LR: 26.32 vs. 3.23) in diagnosis of HCC. CECT produced relatively low IRA, ICA, and diagnostic ability against MR. CONCLUSION: There were substantial variations in liver observations between LI-RADS and LS. Further study is needed to investigate ICA between CECT and MR.

Multidetector computed tomography of superior mesenteric artery: anatomy and pathologies.

The recent introduction of multidetector computed tomography scanners has significantly improved computed tomography angiographic (CTA) applications, especially for the evaluation of medium- and small-arterial structures. CTA of the superior mesenteric artery has been reported previously. However, there have been few systematic and detailed reviews of the superior mesenteric artery pathologies that use CTA. The purpose of this pictorial essay is mainly to review the various superior mesenteric artery pathologies at CTA with our own experiences.

[Clinical value of multi-detector computed tomography source image on detecting extracoronary abnormalities].

OBJECTIVE: To assess the clinical value of multi-detector computed tomography (MDCT) source image on detecting extracoronary abnormalities in a large cohort of patients with suspected coronary artery disease (CAD). METHOD: MDCT source images from 3240 consecutive patients (mean 64.5 years) with suspected CAD were reviewed retrospectively by 2 readers.Extra-coronary findings were classified according to involved organ and level of clinical significance.Following organs were examined:lungs, upper abdomen, spine, chest wall, mediastinum and vascularatures. Clinical relevance of extracoronary findings was considered as either "significant" or "non-significant"."Significant" findings were subclassified as score 1:findings necessitating immediate therapeutic actions, or score 2:findings with uncertain clinical or prognostic relevance, requiring clinical awareness, follow-up or further investigations (non-urgent)."Non-significant" findings were assigned to score 3:findings without clinical implication. The irrelevant incidental findings (e.g. spinal degenerative changes, aortic calcification) were not analyzed. RESULTS: Extracoronary findings was evidenced in 330 patients with 424 abnormalities, 20.3% (67/330) patients had multiple lesions, 16.5% lesions were located in the lungs, 13.2% lesions found in the upper abdomen, 56.8% (241/424) lesions evidenced in the mediastinum,0.9% (4/424) lesions seen in the spine and chest wall, 13.9% (53/424) lesions were related to other vascular disease. Pleural effusion accounts for 5.5% of the mediastinum lesions.Incidence of heart cavity enlargement, heart valve disease, pericardial effusion/calcification, atrial/ventricular perfusion defects, myocardial disease, congenital heart disease, ventricular aneurysm was 14.7% (56/380), 15.5% (59/380), 10.8% (41/380), 3.9% (15/380), 0.8% (3/380), 1.6% (6/380), and 1.8% (7/380) respectively. The clinical significance score 1-3 was 8.5% (36/424) , 81.1% (344/424) , and 10.4% (44/424) respectively.Incidence of detected extracoronary findings was the highest by bone window and the lowest by lung window.Incidence of extracoronary findings was not related to CAD (χ2 = 81.76, C = 0.155, P > 0.05).Inter-reader agreement on extracoronary findings was excellent (Kappa = 0.934, P > 0.05). CONCLUSION: Our data show that it is of clinical value to observe and report extracoronary findings with source image of cardiac MDCT.