Comparison of Four Diagnostic Guidelines for Hepatocellular Carcinoma Using Gadoxetic Acid-enhanced Liver MRI.

Background Noninvasive diagnostic guidelines for hepatocellular carcinoma (HCC) vary across different global geographic areas, especially regarding criteria about gadoxetic acid-enhanced MRI. Purpose To compare the diagnostic performance of four different international HCC diagnosis guidelines and readers' judgment in diagnosing HCC using gadoxetic acid-enhanced MRI in patients at high risk for HCC. Materials and Methods This retrospective study included patients who had not undergone treatment, were at risk for HCC, and who underwent gadoxetic acid-enhanced MRI from January 2015 to June 2018 from 11 tertiary hospitals in South Korea. Four radiologists independently reviewed focal liver lesions (FLLs) according to four guidelines: American Association for the Study of Liver Diseases (AASLD)/Liver Imaging Reporting and Data System (LI-RADS), Korean Liver Cancer Association-National Cancer Center (KLCA-NCC), European Association for the Study of the Liver (EASL), and Asian Pacific Association for the Study of the Liver (APASL). Reader judgment (HCC or not HCC) was also recorded. Malignant FLLs were confirmed at pathology, and histologic and clinical follow-up data were used for benign FLLs. The guidelines' diagnostic performance was compared using generalized estimating equations. Additionally, the diagnostic odds ratio was assessed. Results A total of 2445 FLLs (median size, 27.4 mm) were analyzed in 2237 patients (mean age, 59 years ± 11 [SD]; 1666 male patients); 69.3% (1694 of 2445) were HCCs. KLCA-NCC showed the highest accuracy (80.0%; 95% CI: 78.7, 81.2; P = .001), with high sensitivity in Eastern guidelines (APASL, 89.1% [95% CI: 87.8, 90.3]; KLCA-NCC, 78.2% [95% CI: 76.6, 79.7]) and high specificity in Western guidelines (AASLD/LI-RADS, 89.6% [95% CI: 87.8, 91.2]; EASL, 88.1% [95% CI: 86.2, 89.9]) (P = .001). The diagnostic odds ratios were 20.7 (95% CI: 17.0, 25.3) for AASLD/LI-RADS, 18.9 (95% CI: 15.8, 22.6) for KLCA-NCC, 16.8 (95% CI: 13.8, 20.4) for EASL, and 8.9 (95% CI: 7.4, 10.7) for APASL. The readers' judgment demonstrated higher accuracy than that of the guidelines (accuracy, 86.0%; 95% CI: 84.9, 86.9; P = .001). Conclusion Among four different international HCC diagnosis guidelines, Eastern guidelines demonstrated higher sensitivity, whereas Western guidelines displayed higher specificity. KLCA-NCC achieved the highest accuracy, and AASLD/LI-RADS exhibited the highest diagnostic odds ratio. © RSNA, 2024 Supplemental material is available for this article.

Development and multicenter validation of deep convolutional neural network-based detection of colorectal cancer on abdominal CT.

OBJECTIVES: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network. METHODS: This retrospective study included consecutive patients with colorectal adenocarcinoma who underwent abdominal CT before CRC resection surgery (training set = 379, test set = 103). We customized the 3D U-Net of nnU-Net (CUNET) for CRC detection, which was trained with fivefold cross-validation using annotated CT images. CUNET was validated using datasets covering various clinical situations and institutions: an internal test set (n = 103), internal patients with CRC first determined by CT (n = 54) and asymptomatic CRC (n = 51), and an external validation set from two institutions (n = 60). During each validation, data from the healthy population were added (internal = 60; external = 130). CUNET was compared with other deep CNNs: residual U-Net and EfficientDet. The CAD performances were evaluated using per-CRC sensitivity (true positive/all CRCs), free-response receiver operating characteristic (FROC), and jackknife alternative FROC (JAFROC) curves. RESULTS: CUNET showed a higher maximum per-CRC sensitivity than residual U-Net and EfficientDet (internal test set 91.3% vs. 61.2%, and 64.1%). The per-CRC sensitivity of CUNET at false-positive rates of 3.0 was as follows: internal CRC determined by CT, 89.3%; internal asymptomatic CRC, 87.3%; and external validation, 89.6%. CUNET detected 69.2% (9/13) of CRCs missed by radiologists and 89.7% (252/281) of CRCs from all validation sets. CONCLUSIONS: CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions. KEY POINTS: • Customized 3D U-Net of nnU-Net (CUNET) can be applied to the opportunistic detection of colorectal cancer (CRC) in abdominal CT, helping radiologists detect unexpected CRC. • CUNET showed the best performance at false-positive rates ≥ 3.0, and 30.1% of false-positives were in the colorectum. CUNET detected 69.2% (9/13) of CRCs missed by radiologists and 87.3% (48/55) of asymptomatic CRCs. • CUNET detected CRCs in multiple validation sets composed of varying clinical situations and from different institutions, and CUNET detected 89.7% (252/281) of CRCs from all validation sets.

Development and validation of CT­based radiomics model of PET-negative residual CT masses: a potential biomarker for predicting relapse­free survival in non-Hodgkin lymphoma patients showing complete metabolic response.

PURPOSE: PET-negative residual CT masses (PnRCMs) are usually dismissed as nonviable post-treatment lesions in non-Hodgkin lymphoma (NHL) patients showing complete metabolic response (CMR). We aimed to develop and validate computed tomography (CT)-based radiomics model of PET-negative residual CT mass (PnRCM) for predicting relapse-free survival (RFS) in NHL patients showing CMR. METHODS: A total of 224 patients who showed CMR after completing first-line chemotherapy for PET-avid NHL were recruited for model development. Patients with PnRCM were selected in accordance with the Lugano classification. Three-dimensional segmentation was done by two readers. Radiomic scores (RS) were constructed using features extracted using the Least-absolute shrinkage and selection operator analysis among radiomics features of PnRCMs showing more than substantial interobserver agreement (> 0.6). Cox regression analysis was performed with clinical and radiologic features. The performance of the model was evaluated using area under the curve (AUC). For validation, 153 patients from an outside hospital were recruited and analyzed in the same way. RESULTS: In the model development cohort, 68 (30.4%) patients had PnRCM. Kaplan-Meier analysis showed that patients with PnRCM had significantly (p = 0.005) shorter RFS than those without PnRCM. In Kaplan-Meier analysis, the high RS group showed significantly (p = 0.038) shorter RFS than the low-scoring group. Multivariate Cox regression analysis showed that high IPI score [hazard ratio (HR) 2.46; p = 0.02], treatment without rituximab (HR 3.821; p = 0.019) were factors associated with shorter RFS. In estimating RFS, combined model in both development and validation cohort showed AUC values of 0.81. CONCLUSION: The combined model that incorporated both clinical parameters and CT-based RS showed good performance in predicting relapse in NHL patients with PnRCM.

Particulate matter 10 induces oxidative stress and apoptosis in rhesus macaques skin fibroblast.

Background: Particulate matter (PM) is a major air pollutant that affects human health worldwide. PM can pass through the skin barrier, thus causing skin diseases such as heat rash, allergic reaction, infection, or inflammation. However, only a few studies have been conducted on the cytotoxic effects of PM exposure on large-scale animals. Therefore, herein, we investigated whether and how PM affects rhesus macaque skin fibroblasts. Methods: Rhesus macaque skin fibroblasts were treated with various concentrations of PM10 (1, 5, 10, 50, and 100 µg/mL) and incubated for 24, 48, and 72 h. Then, cell viability assay, TUNEL assay, and qRT-PCR were performed on the treated cells. Further, the reactive oxygen species, glutathione, and cathepsin B levels were determined. The MTT assay revealed that PM10 (>50 µg/mL) proportionately reduced the cell proliferation rate. Results: PM10 treatment increased TUNEL-positive cell numbers, following the pro-apoptosis-associated genes (CASP3 and BAX) and tumor suppressor gene TP53 were significantly upregulated. PM10 treatment induced reactive oxidative stress. Cathepsin B intensity was increased, whereas GSH intensity was decreased. The mRNA expression levels of antioxidant enzyme-related genes (CAT, GPX1 and GPX3) were significantly upregulated. Furthermore, PM10 reduced the mitochondrial membrane potential. The mRNA expression of mitochondrial complex genes, such as NDUFA1, NDUFA2, NDUFAC2, NDUFS4, and ATP5H were also significantly upregulated. In conclusion, these results showed that PM10 triggers apoptosis and mitochondrial damage, thus inducing ROS accumulation. These findings provide potential information on the cytotoxic effects of PM10 treatment and help to understand the mechanism of air pollution-induced skin diseases.

Machine Learning-Based Magnetic Resonance Radiomics Analysis for Predicting Low- and High-Grade Clear Cell Renal Cell Carcinoma.

PURPOSE: To explore whether high- and low-grade clear cell renal cell carcinomas (ccRCC) can be distinguished using radiomics features extracted from magnetic resonance imaging. METHODS: In this retrospective study, 154 patients with pathologically proven clear ccRCC underwent contrast-enhanced 3 T magnetic resonance imaging and were assigned to the development (n = 122) and test (n = 32) cohorts in a temporal-split setup. A total of 834 radiomics features were extracted from whole-tumor volumes using 3 sequences: T2-weighted imaging (T2WI), diffusion-weighted imaging, and contrast-enhanced T1-weighted imaging. A random forest regressor was used to extract important radiomics features that were subsequently used for model development using the random forest algorithm. Tumor size, apparent diffusion coefficient value, and percentage of tumor-to-renal parenchymal signal intensity drop in the tumors were recorded by 2 radiologists for quantitative analysis. The area under the receiver operating characteristic curve (AUC) was generated to predict ccRCC grade. RESULTS: In the development cohort, the T2WI-based radiomics model demonstrated the highest performance (AUC, 0.82). The T2WI-based radiomics and radiologic feature hybrid model showed AUCs of 0.79 and 0.83, respectively. In the test cohort, the T2WI-based radiomics model achieved an AUC of 0.82. The range of AUCs of the hybrid model of T2WI-based radiomics and radiologic features was 0.73 to 0.80. CONCLUSION: Magnetic resonance imaging-based classifier models using radiomics features and machine learning showed satisfactory diagnostic performance in distinguishing between high- and low-grade ccRCC, thereby serving as a helpful noninvasive tool for predicting ccRCC grade.

Hepatocellular carcinoma pathologic grade prediction using radiomics and machine learning models of gadoxetic acid-enhanced MRI: a two-center study.

PURPOSE: To develop a radiomics-based hepatocellular carcinoma (HCC) grade classifier model based on data from gadoxetic acid-enhanced MRI. METHODS: This retrospective study included 137 patients who underwent hepatectomy for a single HCC and gadoxetic acid-enhanced MRI within 60 days before surgery. HCC grade was categorized as low or high (modified Edmondson-Steiner grade I-II vs. III-IV). We used the hepatobiliary phase (HBP), portal venous phase, T2-weighted image(T2WI), and T1-weighted image(T1WI). From the volume of interest in HCC, 833 radiomic features were extracted. Radiomic and clinical features were selected using a random forest regressor, and the classification model was trained and validated using a random forest classifier and tenfold stratified cross-validation. Eight models were developed using the radiomic features alone or by combining the radiomic and clinical features. Models were validated with internal enrolled data (internal validation) and a dataset (28 patients) at a separate institution (external validation). The area under the curve (AUC) of the validation results was compared using the DeLong test. RESULTS: In internal and external validation, the HBP radiomics-only model showed the highest AUC (internal 0.80 ± 0.09, external 0.70 ± 0.09). In external validation, all models showed lower AUC than those for internal validation, while the T2WI and T1WI models failed to predict the HCC grade (AUC 0.30-0.58) in contrast to the internal validation results (AUC 0.67-0.78). CONCLUSION: The radiomics-based machine learning model from gadoxetic acid-enhanced liver MRI could distinguish between low- and high-grade HCCs. The radiomics-only HBP model showed the best AUC among the eight models, good performance in internal validation, and fair performance in external validation.

Radiomics Analysis of Magnetic Resonance Proton Density Fat Fraction for the Diagnosis of Hepatic Steatosis in Patients With Suspected Non-Alcoholic Fatty Liver Disease.

BACKGROUND: This study aimed to assess the diagnostic feasibility of radiomics analysis based on magnetic resonance (MR)-proton density fat fraction (PDFF) for grading hepatic steatosis in patients with suspected non-alcoholic fatty liver disease (NAFLD). METHODS: This retrospective study included 106 patients with suspected NAFLD who underwent a hepatic parenchymal biopsy. MR-PDFF and MR spectroscopy were performed on all patients using a 3.0-T scanner. Following whole-volume segmentation of the MR-PDFF images, 833 radiomic features were analyzed using a commercial program. Radiologic features were analyzed, including median and mean values of the multiple regions of interest and variable clinical features. A random forest regressor was used to extract the important radiomic, radiologic, and clinical features. The model was trained using 20 repeated 10-fold cross-validations to classify the NAFLD steatosis grade. The area under the receiver operating characteristic curve (AUROC) was evaluated using a classifier to diagnose steatosis grades. RESULTS: The levels of pathological hepatic steatosis were classified as low-grade steatosis (grade, 0-1; n = 82) and high-grade steatosis (grade, 2-3; n = 24). Fifteen important features were extracted from the radiomic analysis, with the three most important being wavelet-LLL neighboring gray tone difference matrix coarseness, original first-order mean, and 90th percentile. The MR spectroscopy mean value was extracted as a more important feature than the MR-PDFF mean or median in radiologic measures. Alanine aminotransferase has been identified as the most important clinical feature. The AUROC of the classifier using radiomics was comparable to that of radiologic measures (0.94 ± 0.09 and 0.96 ± 0.08, respectively). CONCLUSION: MR-PDFF-derived radiomics may provide a comparable alternative for grading hepatic steatosis in patients with suspected NAFLD.

Metastatic Renal Cell Carcinoma Manifesting as a Gastric Polyp on CT: A Case Report and Literature Review.

Gastric metastasis from renal cell carcinoma (RCC) is extremely rare, occurring in 0.2% of all RCC cases. Owing to its low prevalence, metachronous gastric metastasis from RCC may be underdiagnosed, and the imaging findings have not been well-established. Herein we present a case of metastatic RCC manifesting as a gastric polyp in a 70-year-old female along with a literature review on the imaging findings of gastric metastases from RCC. In patients presenting with gastric hyper-enhancing polypoid masses, metastasis from RCC should be considered as a differential diagnosis.

Dual-Layer Spectral CT of Pancreas Ductal Adenocarcinoma: Can Virtual Monoenergetic Images of the Portal Venous Phase Be an Alternative to the Pancreatic-Phase Scan?

Objectives: To determine the performance of virtual monoenergetic images (VMIs) of the portal venous phase (PVP) compared with the pancreatic-phase image for pancreatic ductal adenocarcinoma (PDAC) evaluation. Materials and methods: This retrospective study enrolled 64 patients with PDAC who underwent pancreatic CT with dual-layer spectral CT between February 2018 and January 2020. A polychromatic pancreatic-phase image and VMIs at 40 (VMI40), 55 (VMI55), and 70 keV (VMI70) of the PVP were generated. The tumor-to-pancreas contrast-to-noise ratio (CNR), attenuation difference, peripancreatic vascular signal-to-noise ratio (SNR), and CNR were compared among the four images. Subjective image analysis was performed for tumor conspicuity, heterogeneity, size, and arterial invasion. Results: VMI40 and VMI55 demonstrated higher tumor-to-pancreas CNR, attenuation difference, and higher peripancreatic vascular CNR and SNR than the pancreatic-phase image and VMI70 (p < .001). On subjective analysis, VMI55 showed the best tumor conspicuity. Moreover, the inter-reader agreement for arterial invasion in VMIs from the PVP was not inferior to that in the pancreatic-phase image. Conclusion: For evaluating PDAC, the VMI55 of the PVP was superior to the pancreatic-phase image in terms of tumor conspicuity and peripancreatic vascular enhancement. Therefore, the VMI55 of the PVP could be an alternative to the pancreatic-phase scan in patients suspicious of PDAC.

Computer-aided hepatocellular carcinoma detection on the hepatobiliary phase of gadoxetic acid-enhanced magnetic resonance imaging using a convolutional neural network: Feasibility evaluation with multi-sequence data.

BACKGROUND AND OBJECTIVES: Diagnosis of hepatocellular carcinoma (HCC) on liver MRI needs analysis of multi-sequence images. However, developing computer-aided detection (CAD) for every single sequence requires considerable time and labor for image segmentation. Therefore, we developed CAD for HCC on the hepatobiliary phase (HBP) of gadoxetic acid-enhanced magnetic resonance imaging (MRI) using a convolutional neural network (CNN) and evaluated its feasibility on multi-sequence, multi-unit, and multi-center data. METHODS: Patients who underwent gadoxetic acid-enhanced MRI and surgery for HCC in Korea University Anam Hospital (KUAH) and Korea University Guro Hospital (KUGH) were reviewed. Finally, 170 nodules from 155 consecutive patients from KUAH and 28 nodules from 28 patients randomly selected from KUGH were included. Regions of interests were drawn on the whole HCC volume on HBP, T1-weighted (T1WI), T2-weighted (T2WI), and portal venous phase (PVP) images. The CAD was developed from the HBP images of KUAH using customized-nnUNet and post-processed for false-positive reduction. Internal and external validation of the CAD was performed with HBP, T1WI, T2WI, and PVP of KUAH and KUGH. RESULTS: The figure of merit and recall of the jackknife alternative free-response receiver operating characteristic of the CAD for HBP, T1WI, T2WI, and PVP at false-positive rate 0.5 were (0.87 and 87.0), (0.73 and 73.3), (0.13 and 13.3), and (0.67 and 66.7) in KUAH and (0.86 and 86.0), (0.61 and 53.6), (0.07 and 0.07), and (0.57 and 53.6) in KUGH, respectively. CONCLUSIONS: The CAD for HCC on gadoxetic acid-enhanced MRI developed by CNN from HBP detected HCCs feasibly on HBP, T1WI, and PVP of gadoxetic acid-enhanced MRI obtained from multiple units and centers. This result imply that the CAD developed using single MRI sequence may be applied to other similar sequences and this will reduce labor and time for CAD development in multi-sequence MRI.

Magnetic resonance imaging improves stratification of fibrosis and steatosis in patients with chronic liver disease.

PURPOSE: We aimed to compare the diagnostic accuracy of magnetic resonance imaging (MRI) and transient elastography (TE) in assessing liver fibrosis and steatosis in patients with chronic liver disease (CLD). METHODS: Patients who underwent liver biopsy or liver surgery at two academic hospitals between 2017 and 2021 were retrospectively recruited. The stages of liver fibrosis and steatosis were evaluated using histologic examination. Liver stiffness (LS) was assessed using MR elastography (LSMRE) and TE (LSTE). Liver steatosis was assessed using proton density fat fraction (PDFF) and controlled attenuation parameter (CAP). RESULTS: The mean age of the study population (n = 280) was 53.6 years and male sex predominated (n = 199, 71.1%). Nonalcoholic fatty liver disease was the most prevalent (n = 127, 45.5%), followed by hepatitis B virus (n = 112, 40.0%). Hepatocellular carcinoma was identified in 130 patients (46.4%). The proportions of F0, F1, F2, F3, and F4 fibrosis were 13.2%, 31.1%, 9.6%, 16.4%, and 29.7%, respectively. LSMRE had a significantly greater AUROC value than LSTE for detecting F2-F4 (0.846 vs. 0.781, P = 0.046), whereas LSMRE and LSTE similarly predicted F1-4, F3-4, and F4 (all P > 0.05). The proportions of S0, S1, S2, and S3 steatosis were 34.7%, 49.6%, 12.5%, and 3.2%, respectively. PDFF had significantly greater AUROC values than CAP in predicting S1-3 (0.922 vs. 0.806, P < 0.001) and S2-3 (0.924 vs. 0.795, P = 0.005); however, PDFF and CAP similarly predicted S3 (P = 0.086). CONCLUSION: MRI exhibited significantly higher diagnostic accuracy than TE for detecting significant fibrosis and mild or moderate steatosis in patients with CLD.

Diagnostic Feasibility of Magnetic Resonance Elastography Radiomics Analysis for the Assessment of Hepatic Fibrosis in Patients With Nonalcoholic Fatty Liver Disease.

OBJECTIVE: The aim of the study was to investigate the diagnostic feasibility of radiomics analysis using magnetic resonance elastography (MRE) to assess hepatic fibrosis in patients with nonalcoholic fatty liver disease (NAFLD). METHODS: One hundred patients with suspected NAFLD were retrospectively enrolled. All patients underwent a liver parenchymal biopsy. Magnetic resonance elastography was performed using a 3.0-T scanner. After multislice segmentation of MRE images, 834 radiomic features were analyzed using a commercial program. Radiologic features, such as median and mean values of the regions of interest and variable clinical features, were analyzed. A random forest regressor was used to extract important radiomic, radiological, and clinical features. A random forest classifier model was trained to use these features to classify the fibrosis stage. The area under the receiver operating characteristic curve was evaluated using a classifier for fibrosis stage diagnosis. RESULTS: The pathological hepatic fibrosis stage was classified as low-grade fibrosis (stages F0-F1, n = 82) or clinically significant fibrosis (stages F2-F4, n = 18). Eight important features were extracted from radiomics analysis, with the 2 most important being wavelet-high high low gray level dependence matrix dependence nonuniformity-normalized and wavelet-high high low gray level dependence matrix dependence entropy. The median value of the multiple small regions of interest was identified as the most important radiologic feature. Platelet count has been identified as an important clinical feature. The area under the receiver operating characteristic curve of the classifier using radiomics was comparable with that of radiologic measures (0.97 ± 0.07 and 0.96 ± 0.06, respectively). CONCLUSIONS: Magnetic resonance elastography radiomics analysis provides diagnostic performance comparable with conventional MRE analysis for the assessment of clinically significant hepatic fibrosis in patients with NAFLD.

Accuracy of MRI for predicting anterior peritoneal reflection involvement in locally advanced rectal cancer: a comparison with operative findings.

PURPOSE: To assess the diagnostic accuracy of preoperative rectal MRI for anterior peritoneal reflection (APR) involvement in rectal cancer through comparison with the surgeon's operative findings. METHODS: This retrospective study was approved by the institutional review board; informed consent was waived. We enrolled 55 consecutive patients with suspected locally advanced mid-to-upper rectal cancer. All patients underwent rectal MRI using a 3T system. APR involvement in rectal cancer was assessed radiologically using a 5-point scale by two independent board-certified abdominal radiologists. The surgeon's evaluation during surgery was regarded as the gold standard for APR involvement. The accuracy of rectal MRI in predicting APR involvement was obtained. RESULTS: Rectal MRI showed good APR identification (rater 1, 92.7%; rater 2, 94.7%). On preoperative rectal MRI, rater 1 diagnosed 19 (34.5%) patients as having APR involvement and rater 2 diagnosed 28 (50.9%) as having APR involvement. There was moderate agreement (κ = 0.602, p < 0.001) between the two raters with regard to the evaluation of APR involvement. During surgery, the surgeon confirmed APR involvement in 13 (23.6%) patients. The sensitivity, specificity, PPV, and NPV of preoperative MRI for APR involvement were 69.2%, 76.2%, 47.4%, and 88.9%, respectively. The diagnostic accuracy of MRI for predicting APR involvement was 74.6%. CONCLUSION: Preoperative rectal MRI provides accurate anatomical information regarding APR involvement with high conspicuity. However, MRI has relatively low sensitivity (< 70%) and a low PPV (< 50%) with regard to the assessment of APR involvement in rectal tumors. Both rater 1 and rater 2 evaluated these images as positive involvement of APR. The patient underwent laparoscopic low anterior resection after preoperative evaluation. However, during surgery, the surgeon evaluated APR involvement as negative, and the final pathologic staging was confirmed as T3.

Active Learning for Efficient Segmentation of Liver with Convolutional Neural Network-Corrected Labeling in Magnetic Resonance Imaging-Derived Proton Density Fat Fraction.

This study aimed to propose an efficient method for self-automated segmentation of the liver using magnetic resonance imaging-derived proton density fat fraction (MRI-PDFF) through deep active learning. We developed an active learning framework for liver segmentation using labeled and unlabeled data in MRI-PDFF. A total of 77 liver samples on MRI-PDFF were obtained from patients with nonalcoholic fatty liver disease. For the training, tuning, and testing of the liver segmentation, the ground truth of 71 (internal) and 6 (external) MRI-PDFF scans for training and testing were verified by an expert reviewer. For 100 randomly selected slices, manual and deep learning (DL) segmentations for visual assessments were classified, ranging from very accurate to mostly accurate. The dice similarity coefficients for each step were 0.69 ± 0.21, 0.85 ± 0.12, and 0.94 ± 0.01, respectively (p-value = 0.1389 between the first step and the second step or p-value = 0.0144 between the first step and the third step for paired t-test), indicating that active learning provides superior performance compared with non-active learning. The biases in the Bland-Altman plots for each step were - 24.22% (from - 82.76 to - 2.70), - 21.29% (from - 59.52 to 3.06), and - 0.67% (from - 10.43 to 4.06). Additionally, there was a fivefold reduction in the required annotation time after the application of active learning (2 min with, and 13 min without, active learning in the first step). The number of very accurate slices for DL (46 slices) was greater than that for manual segmentations (6 slices). Deep active learning enables efficient learning for liver segmentation on a limited MRI-PDFF.

Deep convolution neural networks to differentiate between COVID-19 and other pulmonary abnormalities on chest radiographs: Evaluation using internal and external datasets.

We aimed to evaluate the performance of convolutional neural networks (CNNs) in the classification of coronavirus disease 2019 (COVID-19) disease using normal, pneumonia, and COVID-19 chest radiographs (CXRs). First, we collected 9194 CXRs from open datasets and 58 from the Korea University Anam Hospital (KUAH). The number of normal, pneumonia, and COVID-19 CXRs were 4580, 3884, and 730, respectively. The CXRs obtained from the open dataset were randomly assigned to the training, tuning, and test sets in a 70:10:20 ratio. For external validation, the KUAH (20 normal, 20 pneumonia, and 18 COVID-19) dataset, verified by radiologists using computed tomography, was used. Subsequently, transfer learning was conducted using DenseNet169, InceptionResNetV2, and Xception to identify COVID-19 using open datasets (internal) and the KUAH dataset (external) with histogram matching. Gradient-weighted class activation mapping was used for the visualization of abnormal patterns in CXRs. The average AUC and accuracy of the multiscale and mixed-COVID-19Net using three CNNs over five folds were (0.99 ± 0.01 and 92.94% ± 0.45%), (0.99 ± 0.01 and 93.12% ± 0.23%), and (0.99 ± 0.01 and 93.57% ± 0.29%), respectively, using the open datasets (internal). Furthermore, these values were (0.75 and 74.14%), (0.72 and 68.97%), and (0.77 and 68.97%), respectively, for the best model among the fivefold cross-validation with the KUAH dataset (external) using domain adaptation. The various state-of-the-art models trained on open datasets show satisfactory performance for clinical interpretation. Furthermore, the domain adaptation for external datasets was found to be important for detecting COVID-19 as well as other diseases.

How to accurately measure the distance from the anal verge to rectal cancer on MRI: a prospective study using anal verge markers.

PURPOSE: To determine an accurate method for localizing rectal cancer using the distance from the anal verge on preoperative MRI. METHODS: This prospective study included 50 patients scheduled for MRI evaluation of rectal cancer. After rectal filling with gel, MRI was performed with two markers attached at the anal verge. The distance between the tumor and the anal verge on a sagittal T2-weighted image (T2WI) was measured independently by two radiologists using six methods divided into three groups of similar measurement approaches, and compared to those obtained on rigid sigmoidoscopy. The anal verge location relative to the external anal sphincter was assessed on oblique coronal T2WI in reference to the markers. Correlation analysis was performed using the intraclass correlation coefficient (ICC) for verification, and a paired t test was used to evaluate the mean differences. RESULTS: The highest correlation (ICC 0.797-0.815) and the least mean difference (0.74-0.85 cm) with rigid sigmoidoscopy, and the least standard deviation (3.12-3.17 cm) were obtained in the direct methods group using a straight line from the anal verge to the tumor. The anal verge was localized within a range of - 1.4 to 1.5 cm (mean - 0.31 cm and - 0.22 cm) from the lower end of the external anal sphincter. CONCLUSION: The direct methods group provided the most accurate tumor distance among the groups. Among the direct methods, we recommend the direct mass method for its simplicity. Despite minor differences in location, the lower end of the external anal sphincter was a reliable anatomical landmark for the anal verge.

Radiologic-Pathologic Correlation of Hepatobiliary Phase Hypointense Nodules without Arterial Phase Hyperenhancement at Gadoxetic Acid-enhanced MRI: A Multicenter Study.

Background Hepatobiliary phase (HBP) hypointense nodules without arterial phase hyperenhancement (APHE) at gadoxetic acid-enhanced MRI may indicate hepatocellular carcinoma (HCC) or nonmalignant cirrhosis-associated nodules. Purpose To assess the distribution of pathologic diagnoses of HBP hypointense nodules without APHE at gadoxetic acid-enhanced MRI and to evaluate clinical and imaging features in differentiating their histologic grades. Materials and Methods This retrospective multicenter study included pathologic analysis-confirmed HBP hypointense nodules without APHE (≤30 mm) in patients with chronic liver disease or cirrhosis screened between January 2008 and June 2016. Central pathologic review by 10 pathologists determined final histologic grades as progressed HCC, early HCC, high-grade dysplastic nodule (DN), and low-grade DN or regenerative nodule. Gadoxetic acid-enhanced MRI features were analyzed by three radiologists. Multivariable logistic regression analyses with elastic net regularization were performed to identify clinical and imaging features for differentiating histologic grades. Results There were 298 patients (mean age, 59 years ± 10; 226 men) with 334 nodules evaluated, and progressed HCCs were diagnosed in 44.0% (147 of 334), early HCCs in 20.4% (68 of 334), high-grade DNs in 27.5% (92 of 334), and low-grade DNs or regenerative nodules in 8.1% (27 of 334). Serum α-fetoprotein level 100 ng/mL or greater (odds ratio, 2.7; P = .01) and MRI features including well-defined margin (odds ratio, 5.5; P = .003), hypointensity at precontrast T1-weighted imaging (odds ratio, 3.2; P < .001), intermediate hyperintensity at T2-weighted imaging (odds ratio, 3.4; P < .001), and restricted diffusion (odds ratio, 1.9; P = .04) were independent predictors for progressed HCC at multivariable analysis. Conclusion In patients at high risk for hepatocellular carcinoma (HCC), hepatobiliary phase hypointense nodules without arterial phase hyperenhancement at gadoxetic acid-enhanced MRI corresponded mainly to progressed HCCs, early HCCs, and high-grade dysplastic nodules. High α-fetoprotein level and some imaging features at MRI helped to differentiate progressed HCC from lower grade nodules. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Motosugi in this issue.

Efficacy of ZOOMit coronal diffusion-weighted imaging and MR texture analysis for differentiating between benign and malignant distal bile duct strictures.

PURPOSE: To investigate the diagnostic efficacy of ZOOMit coronal diffusion-weighted imaging (Z-DWI) and MR texture analysis (MRTA) for differentiating benign from malignant distal bile duct strictures. METHODS: We retrospectively enrolled a total of 71 patients with distal bile duct stricture who underwent magnetic resonance cholangiopancreatography (MRCP). For quantitative analysis, the average apparent diffusion coefficient (ADC) value at suspected stricture sites was assessed on both Z-DWI and conventional DWI (C-DWI). For qualitative analysis, two reviewers independently reviewed two image sets containing different diffusion-weighted images, and receiver operating characteristic (ROC) curve analysis was performed. Several MRTA parameters were extracted from the area of the stricture on the ADC map of the ZOOMit coronal diffusion-weighted images using commercially available software. RESULTS: Among 71 patients, 26 patients were diagnosed with malignant stricture. On quantitative analysis, the average ADC value of the malignant and benign strictures, using Z-DWI, was 1.124 × 10-3 mm2/s and 1.522 × 10-3 mm2/s, respectively (P < 0.001). The average ADC value of the malignant and benign strictures, using C-DWI, was 1.107 × 10-3 mm2/s and 1.519 × 10-3 mm2/s, respectively (P < 0.001). On qualitative analysis, for each reviewer, the area under the ROC curve (AUC) values for differentiating benign from malignant stricture was 0.928 and 0.939, respectively, for the ZOOMit diffusion set and 0.851 and 0.824, respectively, for the conventional diffusion set. Multiple MRTA parameters showed a significantly different distribution for the benign and malignant strictures, including mean, entropy, mean of positive pixels, and kurtosis at spatial filtration values of 0, 5, and 6 mm. CONCLUSION: The addition of Z-DWI to conventional MRCP is helpful in differentiating benign from malignant bile duct strictures, and some MRTA parameters also can be helpful in differentiating benign from malignant distal bile duct strictures.

Predicting cannulation difficulty in endoscopic retrograde cholangiopancreatography using CT image findings: a decision-tree analysis.

BACKGROUND: Difficult cannulation during endoscopic retrograde cholangiopancreatography (ERCP) is associated with increased complications; therefore, its prediction is important. PURPOSE: To identify radiologic risk factors of difficult cannulation during ERCP based on computed tomography (CT) findings and to develop a predictive model for a difficult cannulation. MATERIAL AND METHODS: A total of 171 patients with native papilla who underwent both enhanced CT and ERCP were recruited. Two radiologists independently measured the distal common bile duct (CBD) diameter and choledochoduodenal (CD) angle and analyzed CT images for presence of CBD stone and papilla bulging, size and type of periampullary diverticulum (PAD), and duodenal segment in which major papilla was located. Multivariate logistic regression analysis and decision-tree analysis were performed to identify risk factors for difficult cannulation. RESULTS: Thirty-nine patients underwent a difficult cannulation. The multivariate logistic regression analysis revealed that a smaller CBD diameter, presence of papilla bulging, location of the major papilla other than the descending duodenum, a smaller CD angle, and a higher worrisome PAD score were statistically relevant factors for difficult cannulation (P < 0.049). In the decision-tree analysis, a higher worrisome PAD score was the strongest predictor of difficult cannulation, followed by the presence of papilla bulging, smaller CD angle, and a smaller CBD diameter. The predictive model had an 82.5% overall predictive accuracy. CONCLUSION: The CT findings-based decision-tree analysis model showed a high accuracy in predicting cannulation difficulty and may be helpful for making pre-ERCP strategy.

Three-Dimensional Convolutional Neural Network for Prostate MRI Segmentation and Comparison of Prostate Volume Measurements by Use of Artificial Neural Network and Ellipsoid Formula.

OBJECTIVE. The purposes of this study were to assess the performance of a 3D convolutional neural network (CNN) for automatic segmentation of prostates on MR images and to compare the volume estimates from the 3D CNN with those of the ellipsoid formula. MATERIALS AND METHODS. The study included 330 MR image sets that were divided into 260 training sets and 70 test sets for automated segmentation of the entire prostate. Among these, 162 training sets and 50 test sets were used for transition zone segmentation. Assisted by manual segmentation by two radiologists, the following values were obtained: estimates of ground-truth volume (VGT), software-derived volume (VSW), mean of VGT and VSW (VAV), and automatically generated volume from the 3D CNN (VNET). These values were compared with the volume calculated with the ellipsoid formula (VEL). RESULTS. The Dice similarity coefficient for the entire prostate was 87.12% and for the transition zone was 76.48%. There was no significant difference between VNET and VAV (p = 0.689) in the test sets of the entire prostate, whereas a significant difference was found between VEL and VAV (p < 0.001). No significant difference was found among the volume estimates in the test sets of the transition zone. Overall intraclass correlation coefficients between the volume estimates were excellent (0.887-0.995). In the test sets of entire prostate, the mean error between VGT and VNET (2.5) was smaller than that between VGT and VEL (3.3). CONCLUSION. The fully automated network studied provides reliable volume estimates of the entire prostate compared with those obtained with the ellipsoid formula. Fast and accurate volume measurement by use of the 3D CNN may help clinicians evaluate prostate disease.