Contrast-enhanced CT Exams Research Articles

Comparing fully automated AI body composition measures derived from thin and thick slice CT image data.

To compare fully automated artificial intelligence body composition measures derived from thin (1.25mm) and thick (5mm) slice abdominal CT data. In this retrospective study, fully automated CT-based body composition algorithms for quantifying bone attenuation, muscle attenuation, muscle area, liver attenuation, liver volume, spleen volume, visceral-to-subcutaneous fat ratio (VSR) and aortic calcium were applied to both thin (1.25 × 0.625mm) and thick (5 × 3mm) abdominal CT series from two patient cohorts: unenhanced scans in asymptomatic adults undergoing colorectal cancer screening, and post-contrast scans in patients with colorectal cancer. Body composition measures derived from thin and thick slice data were compared, including correlation coefficients and Bland-Altman analysis. A total of 9882 CT scans (mean age, 57.0years; 4527 women, 5355 men) were evaluated, including 8947 non-contrast and 935 contrast-enhanced CT exams. Very strong positive correlation was observed for all soft tissue measures: muscle attenuation (r2 = 0.97), muscle area (r2 = 0.98), liver attenuation (r2 = 0.99), liver volume (r2 = 0.98) and spleen volume (r2 = 0.99), VSR (r2 = 0.98), and aortic calcium (r2 = 0.92); (p < 0.001 for all). Moderate positive correlation was observed for bone attenuation (r2 = 0.35). Bland-Altman analysis showed strong agreement for muscle attenuation, muscle area, liver attenuation, liver volume and spleen volume. Mean percentage differences amongst body composition measures were less than 5% for VSR (4.6%), muscle area (-0.5%), liver attenuation (0.4%) and liver volume (2.7%) and less than 10% for muscle attenuation (-5.5%) and spleen volume (5.1%). For aortic calcium, thick slice overestimated for Agatston scores between 0 and 100 and > 400 burden in 3.1% and 0.3% relative to thin slice, respectively, but underestimated scores between 100 and 400. Automated body composition measures derived from thin and thick abdominal CT data are strongly correlated and show agreement, particularly for soft tissue applications, making it feasible to use either series for these CT-based body composition algorithms.

Optimizing the Use of Iodinated Contrast Media: Conservation Strategies Used Across Canada During the 2022 Shortage

&#x0D; A shortage of iodinated contrast media (ICM) used in contrast-enhanced CT exams led to the adoption of necessary conservation strategies across Canada.&#x0D; Conservation strategies included multidispensing from single-use and multiuse ICM bottles, diluting or reducing ICM dose volumes, switching to weight-based dosing from fixed-based dosing, lower tube voltage, performing unenhanced CT scans, using alternative imaging modalities, or prioritizing urgent cases. One of the more common alternative conservation strategies was to prioritize urgent cases for contrast-enhanced CT exams.&#x0D; Most medical imaging staff who responded to a national survey on ICM conservation strategies reported they would return to their regular doses used before the shortage despite little to no perceived effect on the contrast conspicuity of images or on patient adverse events with reduced ICM volumes.&#x0D; The ICM shortage represents an opportunity to reconsider ICM usage practices given environmental sustainability concerns with ICM and potential cost savings in reducing its use.&#x0D;

A pipeline for automated deep learning liver segmentation (PADLLS) from contrast enhanced CT exams

Multiple studies have created state-of-the-art liver segmentation models using Deep Convolutional Neural Networks (DCNNs) such as the V-net and H-DenseUnet. Oversegmentation however continues to be a problem. We set forth to address these limitations by developing a an automated workflow that leverages the strengths of different DCNN architectures, resulting in a pipeline that enables fully automated liver segmentation. A Pipeline for Automated Deep Learning Liver Segmentation (PADLLS) was developed and implemented that cascades multiple DCNNs that were trained on more than 200 CT scans. First, a V-net is used to create a rough liver, spleen, and stomach mask. After stomach and spleen pixels are removed using their respective masks and ascites is removed using a morphological algorithm, the scan is passed to a H-DenseUnet to yield the final segmentation. The segmentation accuracy of the pipleline was compared to the H-DenseUnet and the V-net using the SLIVER07 and 3DIRCADb datasets as benchmarks. The PADLLS Dice score for the SLIVER07 dataset was calculated to be 0.957 ± 0.033 and was significantly better than the H-DenseUnet’s score of 0.927 ± 0.044 (p = 0.0219) and the V-net’s score of 0.872 ± 0.121 (p = 0.0067). The PADLLS Dice score for the 3DIRCADb dataset was 0.965 ± 0.016 and was significantly better than the H-DenseUnet’s score of 0.930 ± 0.041 (p = 0.0014) the V-net’s score of 0.874 ± 0.060 (p < 0.001). In conclusion, our pipeline (PADLLS) outperforms existing liver segmentation models, serves as a valuable tool for image-based analysis, and is freely available for download and use.

CT features of acquired cystic kidney disease-associated renal cell carcinoma

PurposeAcquired cystic kidney disease-associated renal cell carcinoma (ACKD-RCC) is a relatively recently described entity with scarce literature describing its imaging appearance (1, 2). The purpose of this study was to determine typical and potentially unique features of ACKD-RCC on CT scan that could aid lesion identification. Materials and methodsA retrospective review of the CT scans of 24 patients with 29 histologically proven ACKD-RCC lesions was performed. Imaging features were recorded based on consensus readings of two radiologists. ResultsTumors ranged in size from 1.2 to 5 cm. Nineteen lesions were right-sided and 10 left-sided. Nineteen lesions were exophytic. One patient had bilateral lesions and three patients had multiple lesions in the same kidney. All lesions had well-defined margins with 21 round, 7 lobulated and 1 crescentic in shape. On non-enhanced exam 4 lesions were hypodense, 16 iso-dense and 9 hyperdense; 5 had gross calcifications. Twenty two patients had contrast-enhanced CT exams, with 13 lesions demonstrating homogeneous enhancement (solid pattern) and 14 having inhomogeneous enhancement (cystic or mixed solid and cystic pattern). Only 1 patient had metastatic disease. Eight patients had a history of renal transplants. ConclusionACKD-RCCs are well-defined lesions of variable size that are almost always rounded and most often exophytic. They occasionally have calcifications and are not uncommonly hyperdense on non-enhanced exam. They are most often iso-dense on non-enhanced exam and can be solid, cystic or mixed in attenuation on enhanced exam.

CT severity indices derived from low monoenergetic images at dual-energy CT may improve prediction of outcome in acute pancreatitis

To determine whether imaging features and severity indices using low monoenergetic DECT images improve diagnostic conspicuity and outcome prediction in acute pancreatitis compared to conventional images. A retrospective study of patients with clinical and radiographic signs of acute pancreatitis who underwent 50 contrast-enhanced CT exams conducted on a single-source DECT was performed. Representative conventional and 50 keV-monoenergetic images were randomized and presented to four abdominal radiologists to determine preferred imaging for detecting fat stranding and parenchymal inflammation. Contrast and signal-to-noise ratios were constructed for necrotic, hypoattenuated, inflamed, and healthy parenchyma. These parameters and the CT severity index (CTSI) were compared between conventional and low monoenergetic images using paired t tests and correlated to clinical outcome. Although preference for conventional images was noted for subtle peri-pancreatic fat stranding (169/200 (85%) reads), there was clear preference for low monoenergetic images among all readers for pancreatic inflammation evaluation (188/200 (94%) reads). Moreover, identification of small, hypoattenuating inflammatory foci on monoenergetic images alone in 13/50 (26%) cases resulted in upstaged CTSI from mild to moderate in 7/50 (14%), associated with longer hospitalization (16 ± 17 days vs. 5 ± 2 days; p < 0.05), ICU admission, and drainage. Quantitatively, a twofold difference between normal and inflamed parenchyma attenuation was identified for monoenergetic (44.8 ± 27.6) vs. conventional (25.1 ± 14.7) images (p < 0.05). Significant increases were seen in the monoenergetic SNR and CNR compared to the conventional images (p < 0.05). DECT low monoenergetic images afford better tissue assessment and demarcation of inflamed pancreatic parenchyma. Additionally, they provide improved characterization of the extent parenchymal necrosis, enabling better classification that may better predict severe clinical outcomes. • DECT low monoenergetic images afford better tissue assessment and demarcation of inflamed pancreatic parenchyma and provide improved characterization of the extent parenchymal necrosis. • Qualitatively, low monoenergetic images were preferred over conventional DECT images for the evaluation of pancreatic inflammation; and quantitatively, there is a twofold difference between normal and inflamed parenchyma attenuation, SNR, and CNR between monoenergetic vs. conventional images. • Monoenergetic imaging identified additional small, hypoattenuating inflammatory foci in 26% resulting in an upstaged CT severity index in 14% associated with longer hospitalization, ICU admission, and drainage, thereby enabling better classification and better prediction of severe clinical outcomes.

Intravenous Contrast Extravasation During CT: A National Data Registry and Practice Quality Improvement Initiative

PurposeEstablish 3 performance benchmarks for intravenous contrast extravasation during CT examinations: extravasation frequency, distribution of extravasation volumes, and severity of injury. Evaluate the effectiveness of implementing practice quality improvement (PQI) methodology in improving performance for these 3 benchmarks. MethodsThe Society of Abdominal Radiology and ACR developed a registry collecting data for contrast extravasation events. The project includes a PQI initiative allowing for process improvement. ResultsAs of December 2013, a total of 58 radiology practices have participated in this project, and 32 practices have completed the 2-cycle PQI. There were a total of 454,497 contrast-enhanced CT exams and 1,085 extravasation events. The average extravasation rate is 0.24%. The median extravasation rate is 0.21%. Most extravasations (82.9%) were between 10 mL and 99 mL. The majority of injuries, 94.6%, are mild in severity, with 4.7% having moderate and 0.8% having severe injuries.Data from practices that completed the PQI process showed a change in the average extravasation rate from 0.28% in the first 6 months to 0.23% in the second 6 months, and the median extravasation rate dropped from 0.25% to 0.16%, neither statistically significant. The distribution of extravasation volumes and the severity of injury did not change between the first and second measurement periods. ConclusionsNational performance benchmarks for contrast extravasation rate, distribution of volumes of extravasate, and distribution of severity of injury are established through this multi-institutional practice registry. The application of PQI failed to have a statistically significant positive impact on any of the 3 benchmarks.

The iPad as a mobile device for CT display and interpretation: diagnostic accuracy for identification of pulmonary embolism

Recent software developments enable interactive, real-time axial, 2D and 3D CT display on an iPad by cloud computing from a server for remote rendering. The purpose of this study was to compare radiologists' interpretative performance on the iPad to interpretation on the conventional picture archive and communication system (PACS). Fifty de-identified contrast-enhanced CT exams performed for suspected pulmonary embolism were compiled as an educational tool to prepare our residents for night call. Two junior radiology attendings blindly interpreted the cases twice, one reader used the PACS first, and the other interpreted on the iPad first. After an interval of at least 2 weeks, the cases were reinterpreted in different order using the other display technique. Sensitivity, specificity, and accuracy for identification of pulmonary embolism were compared for each interpretation method. Pulmonary embolism was present in 25 patients, ranging from main pulmonary artery to subsegmental thrombi. Both readers interpreted 98 % of cases correctly regardless of display platform. There was no significant difference in sensitivity (98 vs 100 %, p = 1.0), specificity (98 vs 96 %, p = 1.0), or accuracy (98 vs 98 %, p = 1.0) for interpretation with the iPad vs the PACS, respectively. CT interpretation on an iPad enabled accurate identification of pulmonary embolism, equivalent to display on the PACS. This mobile device has the potential to expand radiologists' availability for consultation and expedite emergency patient management.