Search

Objectives: To evaluate the feasibility of simulated abbreviated MRI (AMRI) with second shot arterial phase (SSAP) for HCC surveillance and diagnosis. Methods: A total of 129 consecutive patients (age, 58.8 ± 11.4 years; male, 71.3%) underwent gadoxetic acid–enhanced MRI using a modified injection protocol for HCC evaluation from July 2017 to February 2018. The modified injection protocol consisted of routine dynamic imaging (6 mL) and SSAP imaging (4 mL). Two radiologists independently reviewed two AMRI sets: AMRI without SSAP (surveillance set) and AMRI with SSAP (diagnosis set). A modified version of the Liver Imaging Reporting and Data System (LI-RADS) for the diagnosis set was devised by referring to contrast-enhanced ultrasound LI-RADS. Results: Sixty-seven patients with HCC and 62 patients without HCC were included. In the surveillance set, sensitivity and specificity for the detection of patients with HCC were 95.5% and 96.8%, and 94.0% and 96.8% in reviewers 1 and 2, respectively. In the diagnosis set, the scores of most HCCs (76/78, 97.4%) were consistent between LI-RADS of full-protocol and modified LI-RADS of AMRI with SSAP protocol. When the HCC surveillance and diagnosis strategy was changed from strategy 1 (AMRI without SSAP) to strategy 2 (AMRI with SSAP), the recall rate significantly decreased from 52.7 to 3.9% (p < 0.001). Conclusions: The modified LI-RADS score of the AMRI with SSAP protocol showed high agreement with the LI-RADS score of the full protocol. The HCC surveillance and diagnosis strategy using the AMRI with SSAP protocol reduced the recall rate. These results may enable to diagnose HCC simultaneously with surveillance. Key Points: • A modified version of LI-RADS was devised for the diagnostic algorithm using AMRI with the second shot arterial phase (SSAP) by referring to CEUS LI-RADS. • The modified LI-RADS scores using AMRI with SSAP showed a high concordance rate with the conventional LI-RADS score using full-protocol MRI. • The recall rate significantly decreased when the HCC surveillance and diagnosis strategy was changed from strategy 1 (AMRI without SSAP; surveillance then recall test) to strategy 2 (AMRI with SSAP; simultaneous surveillance and diagnosis). [ABSTRACT FROM AUTHOR]

Various ectopic lesions occur in the abdomen and pelvis and affect multiple organs including liver, gallbladder, pancreas, spleen, and organs of the genitourinary system. Ectopic organs may be present outside their normal positions, or ectopic tissues may develop while the original organ exists in its normal position. Both benign and malignant lesions can occur in ectopic organs and tissues. Owing to their unusual location, they can often be misdiagnosed as other lesions or even malignant lesions, such as metastasis or seeding. This multimodality pictorial review provides various cases of ectopic lesions in the abdomen and pelvis, which will help narrow the differential diagnosis and guide clinical decision-making. [ABSTRACT FROM AUTHOR]

Endometrial polypoid lesions encompass various conditions from physiologic changes to benign or malignant disease. Differentiating between the various causes of endometrial polypoid lesions remains difficult by transvaginal sonography. Magnetic resonance imaging (MRI) can provide valuable information regarding endometrial polypoid lesions in situations where it is difficult to obtain histologic samples. Multiparametric MRI including T2-weighted images, T1-weighted fat-saturation contrast-enhanced images, and diffusion-weighted images may be helpful for differentiating the various endometrial polypoid lesions and establishing specific diagnoses and appropriate treatment. [ABSTRACT FROM AUTHOR]

Objectives: Second shot arterial phase (SSAP) imaging is an additional arterial phase image obtained by re-injecting a small amount of contrast medium after routine dynamic imaging in gadoxetic acid-enhanced liver MRI. We aimed to evaluate the feasibility and additional value of a SSAP image in gadoxetic acid-enhanced liver MRI.Methods: One hundred seventy-two patients who underwent SSAP imaging after re-injection of 4 mL of contrast material after routine dynamic imaging (original) in gadoxetic acid-enhanced liver MRIs were included. Motion artifacts on arterial phase (AP) images were rated using a 5-point scale and were compared between the original AP images and SSAP images. We evaluated visual detection rates of arterial hypervascularity on the original AP and SSAP images and their subtraction images in patients with hypervascular hepatocellular carcinoma (HCC).Results: The motion artifact of the SSAP images was significantly lower than that of the original AP images (mean score, 1.76 vs 2.06; p < 0.001). In particular, motion artifacts reduced significantly in the SSAP images of patients with substantial motion artifacts in their original AP images (2.28 vs 3.28; p < 0.001). Among the 30 HCC lesions showing hypervascularity on original AP images, only four (4/30, 13.3%) appeared hyperintense on SSAP images. However, subtraction images of SSAP clearly demonstrated arterial hypervascularity in all HCCs.Conclusion: SSAP images showed significantly fewer motion artifacts than the original AP images. Subtraction images of SSAP maintained the detectability of arterial hypervascularity, although SSAP images showed poor visual detection of arterial hypervascularity of HCC.Key Points: • Arterial phase images obtained after a second injection of a small amount of contrast medium (second shot arterial phase [SSAP]) improved motion artifacts compared to the original AP images. • The motion artifacts improved significantly in the SSAP images of patients with substantial motion artifacts in their original AP images. • Subtraction images of SSAP demonstrated the arterial hypervascularity characteristic of HCC at a level comparable to that of the original AP image. [ABSTRACT FROM AUTHOR]

Purpose: Intraductal papillary neoplasm of the bile duct (IPN-B) is known as a premalignant lesion of invasive cholangiocarcinoma. The purpose of this study was for radiologic-pathologic correlation of morphologic features of IPN-B and to correlate the subclassifications with biological behavior in regard to the bile duct wall invasion. Materials and Methods: A pathologist classified gross morphology of 75 cases (44 men and 31 women, age range, 39-85) of histopathologically proven IPN-B into polypoid, cast-like, superficial-spreading, and cyst-forming type. Preoperative images were retrospectively reviewed by two observers independently and classified the gross appearance of intraductal tumors into the four types. Results: The pathologist classified macroscopic appearances of 75 cases of IPN-B into polypoid type in 26, cast-like intraductal growth in 17, superficial-spreading growth in 21, and cyst-forming type in 11. Two observers classified image findings in accordance with pathologist's classification in 58 and 57 (77% and 76%) among the 75 cases of IPN-B, respectively; 18 and 19 of 26 cases of polypoid type, 14 and 14 of 17 cases of cast-like growth type, 16 and 19 of 21 cases of superficial-spreading type, 10 and 5 of 11 cases of cyst-forming type, respectively. Interobserver agreement for subclassification of tumor morphology was in the category of good agreement (k = 0.651). There was no correlation between morphological subclassification and tendency to invasive cholangiocarcinoma. Conclusion: IPN-Bs can be classified morphologically into polypoid, cast-like growth, superficial-spreading, and cystic type, but there is no correlation between the types and tendency to invasive cholangiocarcinoma. [ABSTRACT FROM AUTHOR]

Objective: To evaluate the outcomes of small (5-10 mm), arterially enhancing nodules (SAENs) shown exclusively at the hepatic arterial phase of CT in a hepatocellular carcinoma (HCC) surveillance population and to determine risk factors for developing HCC.Methods: The study population included 112 patients (male:female = 100:12; aged 36-92 years) with 175 SAENs who were at risk of HCC. We evaluated serial changes during follow-up (1.4-41.8 months, mean 35.7 months) and analysed the initial CT findings of SAENs and the accompanying lesions to elucidate the risk factors for HCC development.Results: Of 175 SAENs, 101(57.7%) disappeared and 34(19.4%) persisted. Forty SAENs (22.9%) became HCC in 33 patients (29.5%). Presence of HCC treatment history (p = 0.005, risk ratio = 7.429), a larger size of SAEN (p = 0.003, risk ratio = 1.630), presence of coexistent HCC (p = 0.021, risk ratio = 3.777) and absence of coexistent typical arterioportal shunts (p = 0.003, risk ratio = 4.459) turned out to be independently significant risk factors for future development of HCC.Conclusion: SAENs were frequently seen in an HCC surveillance population and have a 22.9% probability of becoming HCC on a per-lesion basis. Risk increased particularly when the lesion was associated with a previous or concurrent HCC, a large size or found without a coexistent typical arterioportal shunt. [ABSTRACT FROM AUTHOR]

Objectives: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network.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.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.CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions.• 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.Methods: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network.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.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.CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions.• 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.Results: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network.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.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.CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions.• 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.Conclusions: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network.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.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.CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions.• 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.Key Points: This study aims to develop computer-aided detection (CAD) for colorectal cancer (CRC) using abdominal CT based on a deep convolutional neural network.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.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.CUNET can detect CRC on abdominal CT in patients with various clinical situations and from external institutions.• 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. [ABSTRACT FROM AUTHOR]