Search

The purpose of this study was to determine whether computed tomography (CT) angiography with machine learning (ML) can be used to predict the rapid growth of abdominal aortic aneurysm (AAA).This retrospective study was approved by our institutional review board. Fifty consecutive patients (45 men, 5 women, 73.5 years) with small AAA (38.5 ± 6.2 mm) had undergone CT angiography. To be included, patients required at least 2 CT scans a minimum of 6 months apart. Abdominal aortic aneurysm growth, estimated by change per year, was compared between patients with baseline infrarenal aortic minor axis. For each axial image, major axis of AAA, minor axis of AAA, major axis of lumen without intraluminal thrombi (ILT), minor axis of lumen without ILT, AAA area, lumen area without ILT, ILT area, maximum ILT area, and maximum ILT thickness were measured. We developed a prediction model using an ML method (to predict expansion4 mm/y) and calculated the area under the receiver operating characteristic curve of this model via 10-fold cross-validation.The median aneurysm expansion was 3.0 mm/y. Major axis of AAA and AAA area correlated significantly with future AAA expansion (r = 0.472, 0.416 all P0.01). Machine learning and major axis of AAA were a strong predictor of significant AAA expansion (4 mm/y) (area under the receiver operating characteristic curve were 0.86 and 0.78).Machine learning is an effective method for the prediction of expansion risk of AAA. Abdominal aortic aneurysm area and major axis of AAA are the important factors to reflect AAA expansion.

Objective The aim of this study was to evaluate the effect of cardiac output (CO) on aortic peak enhancement using protocols with different contrast material (CM) injection durations. Methods We used a flow phantom that simulated the human circulatory system. Contrast material was injected at a rate of 4.0 mL/s for a period of 2.5, 5, 10, 15, or 20 seconds for a CO of 2.8, 4.2, and 5.6 L/min. Single-level serial computed tomography scans of the simulated aorta were acquired after the start of CM delivery, and aortic peak enhancement was recorded under the different injection protocols. Results Under a long injection duration protocol (20 seconds), a decrease in CO increased aortic peak enhancement proportionally (CO of 2.8 L/min, 420 Hounsfield units [HU]; CO of 4.2 L/min, 365 HU; CO of 5.6 L/min, 291 HU). However, this effect was decreased under shorter injection duration protocols (5, 10, and 15 seconds); under the shortest (2.5-second) injection duration protocol, a decrease in CO resulted in a decrease in aortic peak enhancement (CO of 2.8 L/min, 36 HU; CO of 4.2 L/min, 51 HU; CO of 5.6 L/min, 55 HU). Conclusions The magnitude of the effect of CO on aortic peak enhancement depends on the CM injection duration.

Objectives The objective of this study was to evaluate the use of 80-kVp scans with knowledge-based iterative model reconstruction (IMR) for computed tomography venography (CTV). Methods This prospective study received institutional review board approval; a previous informed consent was obtained from all participants. We enrolled 30 patients with suspected deep venous thrombosis or pulmonary embolism who were to undergo 80-kVp CTV studies. The images were reconstructed with filtered back projection (FBP), hybrid iterative reconstruction (HIR), and IMR. The venous attenuation, image noise, and contrast-to-noise ratio at the iliac, femoral, and popliteal veins were compared on FBP, HIR, and IMR images. We performed qualitative image analysis (image noise, image contrast, image sharpness, streak artifacts, and overall image quality) of the 3 reconstruction methods and measured their reconstruction times. Results There was no significant difference in venous attenuation among the 3 reconstruction methods (P > 0.05). On IMR images, the image noise was lowest at all 3 venous locations, and the contrast-to-noise ratio was highest. Qualitative evaluation scores were also highest for IMR images. The reconstruction time for FBP, HIR, and IMR imaging was 25.4 ± 1.9 seconds, 43.3 ± 3.3 seconds, and 78.7 ± 6.0 seconds, respectively. Conclusions At clinically acceptable reconstruction times, 80-kVp CTV using the IMR technique yielded better qualitative and quantitative image quality than HIR and FBP.

Objectives The aim of this study was to compare the image quality, the degree of artifacts and the percentage of timing of the optimal hepatic arterial phase (HAP) between test-bolus and bolus-tracking methods on gadolinium–ethoxybenzyl–diethylenetriamine pentaacetic acid (Gd-EOB-DTPA)–enhanced magnetic resonance imaging (MRI). Methods In this prospective study, 60 patients who underwent 3-dimensional dynamic Gd-EOB-DTPA–enhanced hepatic 3-T MRI were enrolled in this study. We randomly assigned the 30 patients to the bolus-tracking method, and another 30 patients to the test-bolus method. Signal-to-noise ratios of the liver and spleen in HAP were compared in the 2 groups. Two radiologists independently assessed the ratio of optimal timing of HAP and the degree of ringing and motion artifacts of the 2 protocols. Results The signal-to-noise ratios of the liver (24.0 [SD, 6.4] vs 20.4 [SD, 4.0]) and spleen (30.0 [SD, 13.3] vs 23.6 [SD, 9.9]) were significantly higher in the test-bolus protocol than in the bolus-tracking protocol. The ratio of optimal timing was also significantly higher with the test-bolus protocol than with the bolus-tracking protocol (76.7% vs 40.0%). The degree of ringing and motion artifacts of test-bolus protocol was significantly lower than that of the bolus-tracking protocol (P < 0.01). Conclusions The test-bolus protocol in dynamic 3-T MRI can yield better qualitative image quality and more optimal timing of HAP images, while reducing the degree of artifacts compared with the bolus-tracking protocol.

PURPOSE To determine the minimum apparent diffusion coefficient (ADC(min)) values of benign and malignant hepatic lesions based on diffusion-weighted imaging and to compare the diagnostic performance of ADC(min) and mean ADC (ADC(mean)) values for differentiating between benign and malignant tumors of the liver. MATERIALS AND METHODS We retrospectively subjected 240 patients with 195 malignant (hepatocellular carcinoma [HCC], n = 137; metastases, n = 44; cholangiocellular carcinoma [CCC], n = 14) and 45 benign tumors (hemangiomas, n = 37; focal nodular hyperplasia [FNH], n = 8). Both ADC(mean) and ADC(min) were evaluated independently by 2 readers, the sensitivity and specificity for the detection of malignancy were calculated, and receiver operating characteristic (ROC) curves were generated. To determine interobserver agreement, we calculated the Pearson correlation coefficient. RESULTS Mean ADC (×10 mm/s) was 1.19 for malignant (HCC, 1.15; metastasis, 1.23; CCC, 1.51) and 2.01 for benign tumors (hemangioma, 2.09; FNH, 1.52; P < 0.001). Minimum ADC was 0.81 for malignant (HCC, 0.79; metastasis, 0.81; CCC, 0.91) and 1.62 for benign tumors (hemangioma, 1.66; FNH, 1.28; P < 0.001). The sensitivity, specificity, and the calculated area under the ROC curve for diagnosing malignant lesions were 86.2%, 86.7%, and 0.942 (reader 1) and 88.7%, 88.9%, and 0.939 (reader 2) for ADC(mean); they were of 92.3%, 97.8%, and 0.984 (reader 1) and 94.9%, 97.8%, and 0.983 (reader 2) for ADC(min). CONCLUSIONS Mean ADC and ADC(min) were valuable for differentiating between malignant and benign hepatic lesions. The area under the ROC curve of ADC(min) was significant higher than that of ADC(mean).

The purpose of this study was to compare the image quality of turbo spin echo (TSE) diffusion-weighted imaging (DWI) and echo-planar imaging (EPI) of the oral cavity region. This retrospective study included 26 patients who had undergone both TSE- and EPI-DWI. Misregistration of DWI with T2-TSE images was assessed in the oral cavity. We also compared geometric distortion, the signal-to-noise ratio (SNR), contrast, and the apparent diffusion coefficient (ADC) for the tongue parotid gland, and spinal cord. On a 5-point scale, 2 radiologists scored the TSE- and EPI-DWI of each patient for ghost artifacts, image contrast, and overall image quality. Distortion in the phase-encoded direction was significantly lower on TSE- than EPI-DWI. The SNR of the tongue and parotid gland was significantly higher on TSE than EPI-DWI except spinal cord. No significant difference was found in contrast and ADC values (except for the ADC of tongue). TSE-DWI yielded higher qualitative scores for all parameters except image contrast. For the oral cavity region, TSE-DWI was superior to EPI-DWI with respect to distortion-free images and superior image quality.

We evaluated whether donor computed tomography (CT) with a combined technique of lower tube voltage and iterative reconstruction (IR) can provide sufficient preoperative information for liver transplantation.We retrospectively reviewed CT of 113 liver donor candidates. Dynamic contrast-enhanced CT of the liver was performed on the following protocol: protocol A (n = 70), 120-kVp with filtered back projection (FBP); protocol B (n = 43), 100-kVp with IR. To equalize the background covariates, one-to-one propensity-matched analysis was used. We visually compared the score of the hepatic artery (A-score), portal vein (P-score), and hepatic vein (V-score) of the 2 protocols and quantitatively correlated the graft volume obtained by CT volumetry (graft-CTv) under the 2 protocols with the actual graft weight.In total, 39 protocol-A and protocol-B candidates showed comparable preoperative clinical characteristics with propensity matching. For protocols A and B, the A-score was 3.87 ± 0.73 and 4.51 ± 0.56 (P < .01), the P-score was 4.92 ± 0.27 and 5.0 ± 0.0 (P = .07), and the V-score was 4.23 ± 0.78 and 4.82 ± 0.39 (P < .01), respectively. Correlations between the actual graft weight and graft-CTv of protocols A and B were 0.97 and 0.96, respectively.Liver-donor CT imaging under 100-kVp plus IR protocol provides better visualization for vascular structures than that under 120-kVp plus FBP protocol with comparable accuracy for graft-CTv, while lowering radiation exposure by more than 40% and reducing contrast-medium dose by 20%.

To compare the radiation dose and image quality of retrospective electrocardiogram (ECG)-gated cardiac computed tomography (CT) between a 100-kVp protocol, hybrid iterative reconstruction (HIR), and display preset optimization and the 120-kVp protocol.We prospectively enrolled 100 patients w