Application of a deep learning algorithm for three-dimensional T1-weighted gradient-echo imaging of gadoxetic acid-enhanced MRI in patients at a high risk of hepatocellular carcinoma.

PURPOSE: To evaluate the efficacy of a vendor-specific deep learning reconstruction algorithm (DLRA) in enhancing image quality and focal lesion detection using three-dimensional T1-weighted gradient-echo images in gadoxetic acid-enhanced liver magnetic resonance imaging (MRI) in patients at a high risk of hepatocellular carcinoma. MATERIALS AND METHODS: In this retrospective analysis, 83 high-risk patients with hepatocellular carcinoma underwent gadoxetic acid-enhanced liver MRI using a 3-T scanner. Triple arterial phase, high-resolution portal venous phase, and high-resolution hepatobiliary phase images were reconstructed using conventional reconstruction techniques and DLRA (AIRTM Recon DL; GE Healthcare) for subsequent comparison. Image quality and solid focal lesion detection were assessed by three abdominal radiologists and compared between conventional and DL methods. Focal liver lesion detection was evaluated using figures of merit (FOMs) from a jackknife alternative free-response receiver operating characteristic analysis on a per-lesion basis. RESULTS: DLRA-reconstructed images exhibited significantly improved overall image quality, image contrast, lesion conspicuity, vessel conspicuity, and liver edge sharpness and reduced subjective image noise, ringing artifacts, and motion artifacts compared to conventionally reconstructed images (all P < 0.05). Although there was no significant difference in the FOMs of non-cystic focal liver lesions between the conventional and DL methods, DLRA-reconstructed images showed notably higher pooled sensitivity than conventionally reconstructed images (P < 0.05) in all phases and higher detection rates for viable post-treatment HCCs in the arterial and hepatobiliary phases (all P < 0.05). CONCLUSIONS: Implementing DLRA can enhance the image quality in 3D T1-weighted gradient-echo sequences of gadoxetic acid-enhanced liver MRI examinations, leading to improved detection of viable post-treatment HCCs.

Feasibility of a deep learning artificial intelligence model for the diagnosis of pediatric ileocolic intussusception with grayscale ultrasonography.

PURPOSE: This study explored the feasibility of utilizing a deep learning artificial intelligence (AI) model to detect ileocolic intussusception on grayscale ultrasound images. METHODS: This retrospective observational study incorporated ultrasound images of children who underwent emergency ultrasonography for suspected ileocolic intussusception. After excluding video clips, Doppler images, and annotated images, 40,765 images from two tertiary hospitals were included (positive-to-negative ratio: hospital A, 2,775:35,373; hospital B, 140:2,477). Images from hospital A were split into a training set, a tuning set, and an internal test set (ITS) at a ratio of 7:1.5:1.5. Images from hospital B comprised an external test set (ETS). For each image indicating intussusception, two radiologists provided a bounding box as the ground-truth label. If intussusception was suspected in the input image, the model generated a bounding box with a confidence score (0-1) at the estimated lesion location. Average precision (AP) was used to evaluate overall model performance. The performance of practical thresholds for the modelgenerated confidence score, as determined from the ITS, was verified using the ETS. RESULTS: The AP values for the ITS and ETS were 0.952 and 0.936, respectively. Two confidence thresholds, CTopt and CTprecision, were set at 0.557 and 0.790, respectively. For the ETS, the perimage precision and recall were 95.7% and 80.0% with CTopt, and 98.4% and 44.3% with CTprecision. For per-patient diagnosis, the sensitivity and specificity were 100.0% and 97.1% with CTopt, and 100.0% and 99.0% with CTprecision. The average number of false positives per patient was 0.04 with CTopt and 0.01 for CTprecision. CONCLUSION: The feasibility of using an AI model to diagnose ileocolic intussusception on ultrasonography was demonstrated. However, further study involving bias-free data is warranted for robust clinical validation.

Perfluorobutane-Enhanced Ultrasound for Characterization of Hepatocellular Carcinoma From Non-hepatocellular Malignancies or Benignancy: Comparison of Imaging Acquisition Methods.

OBJECTIVE: The aim of the work described here was to evaluate the diagnostic performance of perfluorobutane (PFB)-enhanced ultrasound in differentiating hepatocellular carcinoma (HCC) from non-HCC malignancies and other benign lesions using different acquisition methods. METHODS: This prospective study included 69 patients with solid liver lesions larger than 1 cm who were scheduled for biopsy or radiofrequency ablation between September 2020 and March 2021. Lesion diagnosis was designated by three blinded radiologists after reviewing three different sets of acquired images selected according to the following presumed acquisition methods: (i) method A, acquisition up to 5 min after contrast injection; (ii) method B, acquisition up to 1 min after contrast injection with additional Kupffer phase; and (iii) method C, acquisition up to 5 min after contrast injection with additional Kupffer phase. RESULTS: After excluding 7 technical failures, 62 patients with liver lesions (mean size: 24.2 ± 14.8 mm), which consisted of 7 benign lesions, 37 non-HCC malignancies and 18 HCCs. For the HCC diagnosis, method C had the highest sensitivity (75.9%), followed by method B (72.2%) and method A (68.5%), but failed to exhibit statistical significance (p = 0.12). There was no significant difference with respect to the pooled specificity between the three methods (p = 0.28). Diagnostic accuracy was the highest with method C (87.1%) but failed to exhibit statistical significance (p = 0.24). CONCLUSION: Image acquisition up to 5 min after contrast injection with additional Kupffer phase could potentially result in high accuracy and sensitivity without loss of specificity in diagnosing HCC with PFB-enhanced ultrasound.

Sonazoid-enhanced ultrasonography for noninvasive imaging diagnosis of hepatocellular carcinoma: special emphasis on the 2022 KLCA-NCC guideline.

Contrast-enhanced ultrasonography (CEUS) is a noninvasive imaging modality used to diagnose hepatocellular carcinoma (HCC) based on specific imaging features, without the need for pathologic confirmation. Two types of ultrasound contrast agents are commercially available: pure intravascular agents (such as SonoVue) and Kupffer agents (such as Sonazoid). Major guidelines recognize CEUS as a reliable imaging method for HCC diagnosis, although they differ depending on the contrast agents used. The Korean Liver Cancer Association-National Cancer Center guideline includes CEUS with either SonoVue or Sonazoid as a second-line diagnostic technique. However, Sonazoid-enhanced ultrasound is associated with several unresolved issues. This review provides a comparative overview of these contrast agents regarding pharmacokinetic features, examination protocols, diagnostic criteria for HCC, and potential applications in the HCC diagnostic algorithm.

Clinical feasibility of radiofrequency ablation using novel adjustable separable electrodes with a multipurpose needle for treating small hepatocellular carcinomas: a prospective single center study.

BACKGROUND: The novel separable clustered electrode system with two adjustable active tips (ICAEs) and a fine multipurpose needle (MPN) for in situ temperature monitoring and adjuvant agent injection was developed and validated in an animal study. The purpose of this study was to evaluate the technical efficacy and complication of the novel electrode system for patients who have small HCC. METHODS: In this prospective, single-center clinical trial, ten participants with 14 small (≤ 2 cm, BCLC 0-A) HCCs referred for RFA were enrolled. A novel electrode system consisting of two ICAEs and one MPN with a thermometer and side holes was used for RFA. The RF energy was delivered using a multichannel RF system combining bipolar and switching monopolar modes. Technical success, efficacy, and complications were evaluated on immediate and one-month follow-up CT. RESULTS: Technical success was achieved in 92.9% (13/14) of tumors. One participant withdrew consent after RFA, and technical efficacy was achieved in 91.7% (11/12) of tumors. None showed thermal injury to nontarget organs. All patients were discharged the day after RFA without major complications. The active electrode lengths were adjusted in 60% (6/10) of patients during the procedure to tailor the ablation zone (83.3%, n = 5) or treat two tumors with different sizes (16.7%, n = 1). MPN was capable of continuous temperature monitoring during all ablations (100%, 14/14). CONCLUSIONS: RFA using a novel electrode system showed acceptable technical efficacy and safety in patients with small HCCs. Further comparative studies are needed for the investigation of the system's potential benefits compared to conventional electrodes.

Low dose of contrast agent and low radiation liver computed tomography with deep-learning-based contrast boosting model in participants at high-risk for hepatocellular carcinoma: prospective, randomized, double-blind study.

OBJECTIVE: To investigate the image quality and lesion conspicuity of a deep-learning-based contrast-boosting (DL-CB) algorithm on double-low-dose (DLD) CT of simultaneous reduction of radiation and contrast doses in participants at high-risk for hepatocellular carcinoma (HCC). METHODS: Participants were recruited and underwent four-phase dynamic CT (NCT04722120). They were randomly assigned to either standard-dose (SD) or DLD protocol. All CT images were initially reconstructed using iterative reconstruction, and the images of the DLD protocol were further processed using the DL-CB algorithm (DLD-DL). The primary endpoint was the contrast-to-noise ratio (CNR), the secondary endpoint was qualitative image quality (noise, hepatic lesion, and vessel conspicuity), and the tertiary endpoint was lesion detection rate. The t-test or repeated measures analysis of variance was used for analysis. RESULTS: Sixty-eight participants with 57 focal liver lesions were enrolled (20 with HCC and 37 with benign findings). The DLD protocol had a 19.8% lower radiation dose (DLP, 855.1 ± 254.8 mGy·cm vs. 713.3 ± 94.6 mGy·cm, p = .003) and 27% lower contrast dose (106.9 ± 15.0 mL vs. 77.9 ± 9.4 mL, p < .001) than the SD protocol. The comparative analysis demonstrated that CNR (p < .001) and portal vein conspicuity (p = .002) were significantly higher in the DLD-DL than in the SD protocol. There was no significant difference in lesion detection rate for all lesions (82.7% vs. 73.3%, p = .140) and HCCs (75.7% vs. 70.4%, p = .644) between the SD protocol and DLD-DL. CONCLUSIONS: DL-CB on double-low-dose CT provided improved CNR of the aorta and portal vein without significant impairment of the detection rate of HCC compared to the standard-dose acquisition, even in participants at high risk for HCC. KEY POINTS: • Deep-learning-based contrast-boosting algorithm on double-low-dose CT provided an improved contrast-to-noise ratio compared to standard-dose CT. • The detection rate of focal liver lesions was not significantly differed between standard-dose CT and a deep-learning-based contrast-boosting algorithm on double-low-dose CT. • Double-low-dose CT without a deep-learning algorithm presented lower CNR and worse image quality.

Abbreviated MRI for Secondary Surveillance of Recurrent Hepatocellular Carcinoma After Presumed Curative Treatment.

BACKGROUND: Little is known about the performance of abbreviated MRI (AMRI) for secondary surveillance of recurrent hepatocellular carcinoma (HCC) after curative treatment. PURPOSE: To evaluate the detection performance of AMRI for secondary surveillance of HCC after curative treatment. STUDY TYPE: Retrospective. POPULATION: A total of 243 patients (183 men and 60 women; median age, 65 years) who underwent secondary surveillance for HCC using gadoxetic acid-enhanced MRI after more than 2 year of disease-free period following curative treatment, including surgical resection or radiofrequency ablation (RFA). FIELD STRENGTH/SEQUENCE: A 3.0 T/noncontrast AMRI (NC-AMRI) (T2-weighted fast spin-echo, T1-weighted gradient echo, and diffusion-weighted images), hepatobiliary phase AMRI (HBP-AMRI) (T2-weighted fast spin-echo, diffusion-weighted, and HBP images), and full-sequence MRI ASSESSMENT: Four board-certified radiologists independently reviewed NC-AMRI, HBP-AMRI, and full-sequence MRI sets of each patient for detecting recurrent HCC. STATISTICAL TESTS: Per-lesion sensitivity, per-patient sensitivity and specificity for HCC detection at each set were compared using generalized estimating equation. RESULTS: A total of 42 recurred HCCs were confirmed in the 39 patients. The per-lesion and per-patient sensitivities did not show significant differences among the three image sets for either reviewer (P ≥ 0.358): per-lesion sensitivity: 59.5%-83.3%, 59.5%-85.7%, and 59.5%-83.3%, and per-patient sensitivity: 53.9%-83.3%, 56.4%-85.7%, and 53.9%-83.3% for NC-AMRI, HBP-AMRI, and full-sequence MRI, respectively. Per-lesion pooled sensitivities of NC-AMRI, HBP-AMRI, and full-sequence MRI were 72.6%, 73.2%, and 73.2%, with difference of -0.6% (95% confidence interval: -6.7, 5.5) between NC-AMRI and full-sequence MRI and 0.0% (-6.1, 6.1) between HBP-AMRI and full-sequence MRI. Per-patient specificity was not significantly different among the three image sets for both reviewers (95.6%-97.1%, 95.6%-97.1%, and 97.6%-98.5% for NC-AMRI and HBP-AMRI, respectively; P ≥ 0.117). DATA CONCLUSION: NC-AMRI and HBP-AMRI showed no significant difference in detection performance to that of full-sequence gadoxetic acid-enhanced MRI during secondary surveillance for HCC after more than 2-year disease free interval following curative treatment. Based on its good detection performance, short scan time, and lack of contrast agent-associated risks, NC-AMRI is a promising option for the secondary surveillance of HCC. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 2.

Low-contrast-dose liver CT using low monoenergetic images with deep learning-based denoising for assessing hepatocellular carcinoma: a randomized controlled noninferiority trial.

OBJECTIVE: Low monoenergetic images obtained using noise-reduction techniques may reduce CT contrast media requirements. We aimed to investigate the effectiveness of low-contrast-dose CT using dual-energy CT and deep learning-based denoising (DLD) techniques in patients at high risk of hepatocellular carcinoma (HCC). METHODS: We performed a prospective, randomized controlled noninferiority trial at a tertiary hospital between June 2019 and August 2020 (NCT04027556). Patients at high risk of HCC were randomly assigned (1:1) to the standard-contrast-dose group or low-contrast-dose group, which targeted a 40% reduction in contrast medium dose based on lean body weight. HCC conspicuity on arterial phase images was the primary endpoint with a noninferiority margin of 0.2. Images were independently assessed by three radiologists; model-based iterative reconstruction (MBIR) images of the standard-contrast-dose group and low monoenergetic (50-keV) DLD images of the low-contrast-dose group were compared using a generalized estimating equation. RESULTS: Ninety participants (age 59 ± 10 years; 68 men) were analyzed. Compared with the standard-contrast-dose group (n = 47), 40% less contrast media was used in the low-contrast-dose group (n = 43) (107.0 ± 17.1 mL vs. 64.5 ± 11.3 mL, p < 0.001). In the arterial phase, HCC conspicuity on 50-keV DLD images in the low-contrast-dose group was noninferior to that of MBIR images in the standard-contrast-dose group (2.92 vs. 2.56; difference, 0.36; 95% confidence interval, -0.13 to ∞; p = 0.013). CONCLUSIONS: The contrast dose in liver CT can be reduced by 40% without impairing HCC conspicuity when using 50-keV and DLD techniques. KEY POINTS: • In the arterial phase, hepatocellular carcinoma conspicuity on 50-keV deep learning-based denoising images in the low-contrast-dose group was noninferior to that of model-based iterative reconstruction images in the standard-contrast-dose group. • HCC detection was comparable between 50-keV deep learning-based denoising images in the low-contrast-dose group and model-based iterative reconstruction images in the standard-contrast-dose group.

Added Value of the Sliding Sign on Right Down Decubitus CT for Determining Adjacent Organ Invasion in Patients with Advanced Gastric Cancer.

Purpose: To investigate the added value of right down decubitus (RDD) CT when determining adjacent organ invasion in cases of advanced gastric cancer (AGC). Materials and Methods: A total of 728 patients with pathologically confirmed T4a (pT4a), surgically confirmed T4b (sT4b), or pathologically confirmed T4b (pT4b) AGCs who underwent dedicated stomach-protocol CT, including imaging of the left posterior oblique (LPO) and RDD positions, were included in this study. Two radiologists scored the T stage of AGCs using a 5-point scale on LPO CT with and without RDD CT at 2-week intervals and recorded the presence of "sliding sign" in the tumors and adjacent organs and compared its incidence of appearance. Results: A total of 564 patients (77.4%) were diagnosed with pT4a, whereas 65 (8.9%) and 99 (13.6%) patients were diagnosed with pT4b and sT4b, respectively. When RDD CT was performed additionally, both reviewers deemed that the area under the curve (AUC) for differentiating T4b from T4a increased (p < 0.001). According to both reviewers, the AUC for differentiating T4b with pancreatic invasion from T4a increased in the subgroup analysis (p < 0.050). Interobserver agreement improved from fair to moderate (weighted kappa value, 0.296-0.444). Conclusion: RDD CT provides additional value compared to LPO CT images alone for determining adjacent organ invasion in patients with AGC due to their increased AUC values and improved interobserver agreement.

Evaluation of HASTE T2 weighted image with reduced echo time for detecting focal liver lesions in patients at risk of developing hepatocellular carcinoma.

PURPOSE: To compare the image quality and performance of half-Fourier acquisition single-shot turbo spin echo (HASTE) sequences, using compressed sensing (HASTE-CS) and deep-learning based reconstruction (HASTE-DL) in detecting focal liver lesions (FLLs), to those of T2-weighted image using BLADE sequence (T2WI) in patients at risk of developing hepatocellular carcinoma (HCC). MATERIALS AND METHODS: This retrospective study included patients at risk of developing HCC who underwent liver MRI including HASTE-DL, HASTE-CS, T2WI and DWI between January and June 2020. Three radiologists independently reviewed the image quality along with FLL detection in the three T2-based sequences and DWI. Reference lesion characterization was done using the complete set of MRI sequences according to the Liver Imaging Reporting and Data System (LI-RADS) v2018. RESULTS: A total of 227 patients with 88 of whom had FLLs (n = 194, mean size 11.7 ± 10.9 mm) were included. HASTE-DL yielded the highest overall image quality, followed by HASTE-CS and T2WI (3.4 ± 0.5, 3.1 ± 0.6, 2.4 ± 0.5, respectively, P < 0.001 for all). In the detection of FLLs, HASTE-DL showed significantly higher sensitivity than T2WI (51.5 % vs 43.6 %, P = 0.007) whereas HASTE-CS and T2WI bore respectively little difference (P > 0.017) on per-patient basis. For LR-4, -5, -M lesions, HASTE-DL had significantly higher figure of merit than that of T2WI (0.58 vs 0.52, P < 0.001) in per-lesion basis. CONCLUSION: HASTE-DL demonstrated better image quality and higher performance for FLL detection than conventional T2WI in patients at risk of developing HCC.

Diagnostic Performance of Spin-Echo Echo-Planar Imaging Magnetic Resonance Elastography in 3T System for Noninvasive Assessment of Hepatic Fibrosis.

OBJECTIVE: To validate the performance of 3T spin-echo echo-planar imaging (SE-EPI) magnetic resonance elastography (MRE) for staging hepatic fibrosis in a large population, using surgical specimens as the reference standard. MATERIALS AND METHODS: This retrospective study initially included 310 adults (155 undergoing hepatic resection and 155 undergoing donor hepatectomy) with histopathologic results from surgical liver specimens. They underwent 3T SE-EPI MRE ≤ 3 months prior to surgery. Demographic findings, underlying liver disease, and hepatic fibrosis pathologic stage according to METAVIR were recorded. Liver stiffness (LS) was measured by two radiologists, and inter-reader reproducibility was evaluated using the intraclass correlation coefficient (ICC). The mean LS of each fibrosis stage (F0-F4) was calculated in total and for each etiologic subgroup. Comparisons among subgroups were performed using the Kruskal-Wallis test and Conover post-hoc test. The cutoff values for fibrosis staging were estimated using receiver operating characteristic (ROC) curve analysis. RESULTS: Inter-reader reproducibility was excellent (ICC, 0.98; 95% confidence interval, 0.97-0.99). The mean LS values were 1.91, 2.41, 3.24, and 5.41 kPa in F0-F1 (n = 171), F2 (n = 26), F3 (n = 38), and F4 (n = 72), respectively. The discriminating cutoff values for diagnosing ≥ F2, ≥ F3, and F4 were 2.18, 2.71, and 3.15 kPa, respectively, with the ROC curve areas of 0.97-0.98 (sensitivity 91.2%-95.9%, specificity 90.7%-99.0%). The mean LS was significantly higher in patients with cirrhosis (F4) of nonviral causes, such as primary biliary cirrhosis (9.56 kPa) and alcoholic liver disease (7.17 kPa) than in those with hepatitis B or C cirrhosis (4.28 and 4.92 kPa, respectively). There were no statistically significant differences in LS among the different etiologic subgroups in the F0-F3 stages. CONCLUSION: The 3T SE-EPI MRE demonstrated high interobserver reproducibility, and our criteria for staging hepatic fibrosis showed high diagnostic performance. LS was significantly higher in patients with non-viral cirrhosis than in those with viral cirrhosis.

Image quality in liver CT: low-dose deep learning vs standard-dose model-based iterative reconstructions.

OBJECTIVES: To compare the overall image quality and detectability of significant (malignant and pre-malignant) liver lesions of low-dose liver CT (LDCT, 33.3% dose) using deep learning denoising (DLD) to standard-dose CT (SDCT, 100% dose) using model-based iterative reconstruction (MBIR). METHODS: In this retrospective study, CT images of 80 patients with hepatic focal lesions were included. For noninferiority analysis of overall image quality, a margin of - 0.5 points (scored in a 5-point scale) for the difference between scan protocols was pre-defined. Other quantitative or qualitative image quality assessments were performed. Additionally, detectability of significant liver lesions was compared, with 64 pairs of CT, using the jackknife alternative free-response ROC analysis, with noninferior margin defined by the lower limit of 95% confidence interval (CI) of the difference of figure-of-merit less than - 0.1. RESULTS: The mean overall image quality scores with LDCT and SDCT were 3.77 ± 0.38 and 3.94 ± 0.34, respectively, demonstrating a difference of - 0.17 (95% CI: - 0.21 to - 0.12), which did not cross the predefined noninferiority margin of - 0.5. Furthermore, LDCT showed significantly superior quantitative results of liver lesion contrast to noise ratio (p < 0.05). However, although LDCT scored higher than the average score in qualitative image quality assessments, they were significantly lower than those of SDCT (p < 0.05). Figure-of-merit for lesion detection was 0.859 for LDCT and 0.878 for SDCT, showing noninferiority (difference: - 0.019, 95% CI: - 0.058 to 0.021). CONCLUSION: LDCT using DLD with 67% radiation dose reduction showed non-inferior overall image quality and lesion detectability, compared to SDCT. KEY POINTS: • Low-dose liver CT using deep learning denoising (DLD), at 67% dose reduction, provided non-inferior overall image quality compared to standard-dose CT using model-based iterative reconstruction (MBIR). • Low-dose CT using DLD showed significantly less noise and higher CNR lesion to liver than standard-dose CT using MBIR and demonstrated at least average image quality score among all readers, albeit with lower scores than standard-dose CT using MBIR. • Low-dose liver CT showed noninferior detectability for malignant and pre-malignant liver lesions, compared to standard-dose CT.

The feasibility of deep learning-based synthetic contrast-enhanced CT from nonenhanced CT in emergency department patients with acute abdominal pain.

Our objective was to investigate the feasibility of deep learning-based synthetic contrast-enhanced CT (DL-SCE-CT) from nonenhanced CT (NECT) in patients who visited the emergency department (ED) with acute abdominal pain (AAP). We trained an algorithm generating DL-SCE-CT using NECT with paired precontrast/postcontrast images. For clinical application, 353 patients from three institutions who visited the ED with AAP were included. Six reviewers (experienced radiologists, ER1-3; training radiologists, TR1-3) made diagnostic and disposition decisions using NECT alone and then with NECT and DL-SCE-CT together. The radiologists' confidence in decisions was graded using a 5-point scale. The diagnostic accuracy using DL-SCE-CT improved in three radiologists (50%, P = 0.023, 0.012, < 0.001, especially in 2/3 of TRs). The confidence of diagnosis and disposition improved significantly in five radiologists (83.3%, P < 0.001). Particularly, in subgroups with underlying malignancy and miscellaneous medical conditions (MMCs) and in CT-negative cases, more radiologists reported increased confidence in diagnosis (83.3% [5/6], 100.0% [6/6], and 83.3% [5/6], respectively) and disposition (66.7% [4/6], 83.3% [5/6] and 100% [6/6], respectively). In conclusion, DL-SCE-CT enhances the accuracy and confidence of diagnosis and disposition regarding patients with AAP in the ED, especially for less experienced radiologists, in CT-negative cases, and in certain disease subgroups with underlying malignancy and MMCs.

Abbreviated Magnetic Resonance Imaging With Breath-Hold Three-Dimensional Magnetic Resonance Cholangiopancreatography: Assessment of Malignant Risk of Pancreatic Intraductal Papillary Mucinous Neoplasm.

BACKGROUND: For surveillance of pancreatic intraductal papillary mucinous neoplasms (IPMNs), magnetic resonance imaging (MRI)/magnetic resonance cholangiopancreatography (MRCP) is preferred over computed tomography, but the long acquisition time limits its use. PURPOSE: To investigate the diagnostic performance of abbreviated MRI with breath-hold (BH) three-dimensional MRCP (abbreviated MRI-BH) for malignant risk assessment of pancreatic IPMN. STUDY TYPE: Retrospective. POPULATION: Two hundred and thirty-five patients with IPMNs (M:F = 115:120; mean age ± SD, 66 ± 9 years; typical imaging features with ≥2-year stability [N = 172] and histopathologically confirmed [N = 63]). FIELD STRENGTH/SEQUENCE: 3.0 T/ abbreviated MRI-BH (single-shot fast spin-echo, T1W fat-suppressed gradient-echo sequence, and BH-3D-MRCP). ASSESSMENT: Abbreviated MRI-BH was reviewed by three reviewers, and its diagnostic performance was assessed using the predetermined scoring system. The diagnostic performance for the mural nodule detection was assessed. Additionally, diagnostic performance of abbreviated MRI was compared with that of full-sequence MRI. STATISTICAL TESTS: Area under the receiver operating characteristic curve (AUC) with z-test, and linear-weighted kappa values. RESULTS: Thirty-five patients had malignant IPMN. At a cut-off score ≥3, AUCs of abbreviated MRI-BH for detecting malignant IPMN were 0.959 for reviewer 1, 0.962 for reviewer 2, and 0.956 for reviewer 3. The sensitivity of reviewers 1, 2, and 3 was 97.1% for all, and the specificity was 85.5%, 86.0%, and 85.0%, respectively. Regarding mural nodule detection (N = 22), abbreviated MRI-BH demonstrated a sensitivity of 95.5% and a specificity of 88.3% for reviewer 1, a sensitivity of 86.4% and a specificity of 92.0% for reviewer 2, and a sensitivity of 86.4% and a specificity of 89.2% for reviewer 3. There were no significant differences between AUC of abbreviated MRI-BH and that of full-sequence MRI in the three reviewers (P > 0.05). DATA CONCLUSION: Abbreviated MRI-BH showed good diagnostic performance for detecting malignant IPMNs by using a predetermined scoring system. LEVEL OF EVIDENCE: 3 TECHNICAL EFFICACY STAGE: 2.

Diagnostic Performance of a New Convolutional Neural Network Algorithm for Detecting Developmental Dysplasia of the Hip on Anteroposterior Radiographs.

OBJECTIVE: To evaluate the diagnostic performance of a deep learning algorithm for the automated detection of developmental dysplasia of the hip (DDH) on anteroposterior (AP) radiographs. MATERIALS AND METHODS: Of 2601 hip AP radiographs, 5076 cropped unilateral hip joint images were used to construct a dataset that was further divided into training (80%), validation (10%), or test sets (10%). Three radiologists were asked to label the hip images as normal or DDH. To investigate the diagnostic performance of the deep learning algorithm, we calculated the receiver operating characteristics (ROC), precision-recall curve (PRC) plots, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) and compared them with the performance of radiologists with different levels of experience. RESULTS: The area under the ROC plot generated by the deep learning algorithm and radiologists was 0.988 and 0.988-0.919, respectively. The area under the PRC plot generated by the deep learning algorithm and radiologists was 0.973 and 0.618-0.958, respectively. The sensitivity, specificity, PPV, and NPV of the proposed deep learning algorithm were 98.0, 98.1, 84.5, and 99.8%, respectively. There was no significant difference in the diagnosis of DDH by the algorithm and the radiologist with experience in pediatric radiology (p = 0.180). However, the proposed model showed higher sensitivity, specificity, and PPV, compared to the radiologist without experience in pediatric radiology (p < 0.001). CONCLUSION: The proposed deep learning algorithm provided an accurate diagnosis of DDH on hip radiographs, which was comparable to the diagnosis by an experienced radiologist.

Diagnostic Performance of LI-RADS Treatment Response Algorithm for Hepatocellular Carcinoma: Adding Ancillary Features to MRI Compared with Enhancement Patterns at CT and MRI.

Background The Liver Imaging Reporting and Data System Treatment Response (LR-TR) algorithm is used to assess the response of hepatocellular carcinoma (HCC) to local-regional therapy (LRT) based on enhancement patterns. The potential value of adding MRI ancillary features (AFs) needs to be investigated. Purpose To evaluate the potential application of MRI AFs in category adjustment to detect pathologic tumor viability in comparison with the LR-TR algorithm in CT and gadoxetic acid-enhanced MRI. Materials and Methods This retrospective study included patients with HCCs treated with LRT followed by surgical resection or liver transplantation between January 2014 and December 2017 who underwent both post-LRT CT and gadoxetic acid-enhanced MRI. For each treated observation, treatment response (TR) categories were assigned based on a consensus reading of three radiologists according to the LR-TR algorithm in CT and MRI and according to the MRI-modified TR algorithm in which MRI AFs were allowed for category adjustment. The diagnostic performances of CT LR-TR viable, MRI LR-TR viable, and MRI-modified TR viable categories were compared intraindividually with the McNemar test, with pathologic tumor viability used as a reference standard. Results A total of 138 patients (119 men; mean age, 58 years ± 9 [standard deviation]) with 138 treated observations (108 pathologically viable) were evaluated. The sensitivity and specificity of CT LR-TR viable and MRI LR-TR viable categories for predicting tumor viability were 73% (79 of 108 lesions; 95% confidence interval [CI]: 64%, 81%) versus 76% (82 of 108 lesions; 95% CI: 67%, 84%) and 90% (27 of 30 lesions; 95% CI: 74%, 98%) versus 83% (25 of 30 lesions; 95% CI: 65%, 94%), respectively, without differences between CT and MRI (P = .65 and P = .63, respectively). MRI-modified TR viable category had higher sensitivity (84% [91 of 108 lesions; 95% CI: 76%, 91%]) than CT or MRI LR-TR viable category (P = .002 and P = .01, respectively), without difference in specificity (80% [24 of 30 lesions]; 95% CI: 61%, 92%) (P = .38 and P > .99, respectively). Conclusion The application of MRI ancillary features to the Liver Imaging Reporting and Data System Treatment Response algorithm resulted in higher sensitivity and no change in specificity compared with CT or MRI enhancement patterns alone in the prediction of pathologic tumor viability in patients with hepatocellular carcinoma. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Do and Mendiratta-Lala in this issue.

LI-RADS treatment response categorization on gadoxetic acid-enhanced MRI: diagnostic performance compared to mRECIST and added value of ancillary features.

OBJECTIVES: This study was conducted in order to assess the performance of the Liver Imaging Reporting and Data System (LI-RADS) treatment response (TR) (LR-TR) categorization on gadoxetic acid-enhanced MRI (Gd-EOB-MRI) for detecting viable tumors in hepatocellular carcinoma (HCC) treated with locoregional treatment (LRT) and to investigate the added value of ancillary features (AFs) to conventional enhancement-based criteria. METHODS: This retrospective study included 183 patients with Gd-EOB-MRI after LRT for HCC and appropriate reference standards for tumor viability (84 viable and 99 nonviable). Two independent radiologists assigned per-lesion mRECIST and TR categories (TR-nonviable, TR-equivocal, or TR-viable) according to the LR-TR algorithm and modified LR-TR algorithms including mLR-TR(TP) allowing transitional phase (TP) washout and mLR-TR(AF) allowing category adjustment by applying AFs. Diagnostic performances of imaging criteria were compared using the Cochran's Q test with post hoc analysis. RESULTS: For detecting viable tumors, LR-TR-viable resulted in sensitivities of 64.5%/39.3% and specificities of 98.0%/98.0% in reviewers 1/2. In comparison to LR-TR-viable, mRECIST-viable, mLR-TR(TP)-viable, and mLR-TR(AF)-viable showed significantly higher sensitivities (92.9%/94.0%, 77.4%/56.6%, and 86.9%/83.3% in reviewers 1/2) (ps < 0.001). The specificity of mRECIST-viable (73.7%/62.6%) was significantly lower than that of LR-TR-viable (ps < 0.001), while those of mLR-TR(TP)-viable and mLR-TR(AF)-viable were greater than 95% (98.0%/96.0% and 97.0%/96.0%), statistically equivalent to LR-TR-viable (ps > 0.05). TR-equivocal was least assigned on mLR-TR(AF) (1.1%/7.7%) than LR-TR (15.8%/32.2%) or mLR-TR(TP) (6.6%/23.5%) in both reviewers. CONCLUSION: The LR-TR algorithm on Gd-EOB-MRI provides a specific diagnosis of viable tumor but with limited sensitivity. By applying AFs in the category adjustment, more sensitive and confident diagnosis can be achieved without significant loss of specificity. KEY POINTS: • The LI-RADS treatment response (LR-TR) algorithm on Gd-EOB-MRI provides a highly specific diagnosis of viable HCC but with limited sensitivity. • The inferior sensitivity of LR-TR-viable category to that of mRECIST can be improved by applying ancillary features in the category adjustment.

The Efficacy of Lymph Node Embolization Using N-Butyl Cyanoacrylate Compared to Ethanol Sclerotherapy in the Management of Symptomatic Lymphorrhea after Pelvic Surgery.

PURPOSE: To compare the efficacy of lymph node (LN) embolization using N-butyl cyanoacrylate versus ethanol sclerotherapy in the management of symptomatic postoperative pelvic lymphorrhea. MATERIALS AND METHODS: Thirty-three patients with 40 instances of symptomatic postoperative lymphorrhea were treated with either LN embolization or sclerotherapy at Seoul National University Hospital from January 2009 to July 2017 and were retrospectively included (LN embolization group: 24 lymphoceles of 19 patients, mean age of 59.29 years; sclerotherapy group: 16 lymphoceles of 14 patients, mean age of 60.95 years). The types of operations were hysterectomy and bilateral oophorectomy with pelvic lymph node dissection (n = 9), radical prostatectomy (n = 3), and renal transplantation (n = 2) for the sclerotherapy group and radical prostatectomy (n = 10) and hysterectomy and bilateral oophorectomy with pelvic lymph node dissection (n = 9) for the LN embolization group. The 3 most common indications of treatment were lower extremity edema (n = 11), pain (n = 11), and fever (n = 8). The amount of leak before treatment (initial daily drainage) and clinical outcomes, including the clinical success rate in 3 weeks, treatment period, and complication rate were compared between both groups. RESULTS: LN embolization showed a higher 3-week clinical success rate than sclerotherapy in a univariate analysis (83.3% and 43.8%, P = .026). There was no statistically significant difference in the treatment period and the complication rate (7.1 days and 12.3 days, P = .098; 8.3% and 25.0%, P = .184). CONCLUSIONS: LN embolization is more effective for treating postoperative pelvic lymphorrhea than sclerotherapy with similar safety.

Isolated Main Pancreatic Duct Dilatation: CT Differentiation Between Benign and Malignant Causes.

OBJECTIVE: The purpose of this study is to retrospectively evaluate the differential CT features of isolated benign and malignant main pancreatic duct (MPD) dilatation and to investigate whether the diagnostic performance of radiologists can be improved with knowledge of these differential CT features. MATERIALS AND METHODS: Forty-one patients who had isolated MPD dilatation without any visible mass on CT from January 2000 to October 2016 were retrospectively enrolled in the study. Two radiologists reviewed CT images in consensus for the location, shape (smooth vs abrupt), length of transition, dilated pancreatic duct (PD) diameter, presence of duct penetrating sign, parenchymal atrophy, attenuation difference, associated pancreatitis, calcification, PD or common bile duct (CBD) enhancement, and perilesional cyst. The chi-square test, Fisher exact test, and t test were used to find the differential CT features of benign and malignant MPD dilatation. Two successive review sessions for differentiation between the two disease entities were then independently performed by three other reviewers with differing expertise, with the use of a 5-point confidence scale. The first session provided no information for differentiation; however, reviewers were aware of the results of univariate analyses in the second session. The diagnostic performance of the radiologists was evaluated using a pairwise comparison of ROC curves. RESULTS: A total of 19 benign and 22 malignant MPD dilatations were identified. In patients with benign MPD dilatation, transition areas were frequently located in the head (57.9% [11/19] vs 13.6% [3/22], p = 0.003) and showed significantly shorter (< 6.1 mm) (78.9% [15/19] vs 9.1% [2/22], p < 0.0001) and smooth transition (89.5% [17/19] vs 9.1% [2/22], p < 0.0001). Duct penetrating sign was exclusively observed in patients with benign MPD dilatation (73.7% [14/19] vs 0% [0/22], p < 0.0001). In contrast, malignant MPD dilatation frequently was accompanied by attenuation difference (63.6% [14/22] vs 10.5% [2/19], p = 0.001) and associated PD or CBD enhancement (36.4% [8/22] vs 0% [0/19], p = 0.003). The AUC values of three reviewers significantly increased from 0.653, 0.587, and 0.884 to 0.864, 0.964, and 0.908, respectively, with knowledge of significant CT features (p = 0.013, p < 0.0001, and p = 0.701, respectively). CONCLUSION: Distal, long (≥ 6.1 mm), and abrupt transition, the absence of duct penetrating sign, and the presence of attenuation difference and PD or CBD enhancement were highly suggestive CT findings for differentiation of malignant from benign MPD dilatation. The diagnostic performance of radiologists with regard to differentiation was significantly improved with knowledge of these highly suggestive CT criteria.