Deep learning on nonenhanced cardiac MRI data can detect the presence and extent of chronic myocardial infarction. This approach may have potential to reduce use of gadolinium contrast administration. Background Renal impairment is common in patients with coronary artery disease and, if severe, late gadolinium enhancement (LGE) imaging for myocardial infarction (MI) evaluation cannot be performed. Purpose To develop a fully automatic framework for chronic MI delineation via deep learning on non–contrast material–enhanced cardiac cine MRI. Materials and Methods In this retrospective single-center study, a deep learning model was developed to extract motion features from the left ventricle and delineate MI regions on nonenhanced cardiac cine MRI collected between October 2015 and March 2017. Patients with chronic MI, as well as healthy control patients, had both nonenhanced cardiac cine (25 phases per cardiac cycle) and LGE MRI examinations. Eighty percent of MRI examinations were used for the training data set and 20% for the independent testing data set. Chronic MI regions on LGE MRI were defined as ground truth. Diagnostic performance was assessed by analysis of the area under the receiver operating c...

To assess the image quality of aorta obtained by dual-source computed tomography angiography (DSCTA), performed with high pitch, low tube voltage, and low iodine concentration contrast medium (CM) with images reconstructed using iterative reconstruction (IR). One hundred patients randomly allocated to receive one of two types of CM underwent DSCTA with the electrocardiogram-triggered Flash protocol. In the low-iodine group, 50 patients received CM containing 270 mg I/mL and were scanned at low tube voltage (100 kVp). In the high-iodine CM group, 50 patients received CM containing 370 mg I/mL and were scanned at the tube voltage (120 kVp). The filtered back projection (FBP) algorithm was used for reconstruction in both groups. In addition, the IR algorithm was used in the low-iodine group. Image quality of the aorta was analyzed subjectively by a 3-point grading scale and objectively by measuring the CT attenuation in terms of the signal- and contrast-to-noise ratios (SNR and CNR, respectively). Radiation and CM doses were compared. The CT attenuation, subjective image quality assessment, SNR, and CNR of various aortic regions of interest did not differ significantly between two groups. In the low-iodine group, images reconstructed by FBP and IR demonstrated significant differences in image noise, SNR, and CNR (p<0.05). The low-iodine group resulted in 34.3% less radiation (4.4 ± 0.5 mSv) than the high-iodine group (6.7 ± 0.6 mSv), and 27.3% less iodine weight (20.36 ± 2.65 g) than the high-iodine group (28 ± 1.98 g). Observers exhibited excellent agreement on the aortic image quality scores (κ = 0.904). CT images of aorta could be obtained within 2 s by using a DSCT Flash protocol with low tube voltage, IR, and low-iodine-concentration CM. Appropriate contrast enhancement was achieved while maintaining good image quality and decreasing the radiation and iodine doses.

This study aims to explore computed tomography (CT) and magnetic resonance imaging (MRI) features of patients diagnosed with primary cardiac angiosarcoma. The study involved the analysis of 12 patients diagnosed with primary cardiac angiosarcoma who underwent non-contrast (8/12) or contrast-enhanced CT (10/12) or MRI (4/12). Imaging appearances, including the tumor location and adjacent infiltration, were analyzed. All 12 lesions were located in the right atrium with a broad base. Adjacent invasion including the tricuspid valve and right ventricle (2/12), inferior or superior vena cava (2/12), pericardium (10/12), and right coronary artery (7/12) was common. On unenhanced CT scans, tumors in two patients were homogeneous in density, whereas the others were inhomogeneous. Ten patients showed heterogeneous enhancement. The enhancement pattern showed no direct correlation with the differentiation degree of the tumor. Four lesions manifested as heterogeneous intensity, with hyperintense hemorrhage foci on both T1- and T2-weighted MRI. Three showed rim enhancement and one showed patchy enhancement. Primary cardiac angiosarcoma often involves the right side of the heart with infiltration of peripheral structures. CT features include typical inhomogeneous density on unenhanced scans and heterogeneous centripetal enhancement on enhanced scans. A cauliflower-like appearance on both T1- and T2-weighted MRI is common. The characteristic enhancement pattern of MRI remains to be determined.

The purpose of this study is to evaluate the hemodynamic characteristics of the true lumen (TL) and the false lumen (FL) in 16 patients with aortic dissection (AD) using 4D flow magnetic resonance imaging (MRI) and thoracic and abdominal computed tomography (CT) angiography. The quantitative parameters that were measured in the TL and FL included velocity and flow. The mean area and regurgitant fraction of the TL were significantly lesser at all four levels (p < 0.05); the average through-plane velocity, peak velocity magnitude, average net flow, peak flow, and net forward volume in the TL were considerably higher (p < 0.05). The intimal entry’s size was negatively correlated with the blood flow velocity and flow rate in the TL (p < 0.05) and positively correlated with the average through-plane velocity, average net flow, and peak flow in the FL (p < 0.05); the blood flow indices in the TL were enhanced with an increase in the intimal entry numbers (p < 0.05) and the peak flow in the FL was lowered (p = 0.025); if FL thrombosis existed, the average through-plane velocity and peak velocity magnitude in the TL were substantially higher (p < 0.05). 4D flow MRI facilitates qualitative and quantitative analysis of the alterations in the abdominal aortic blood flow patterns.