The impact of COVID-19 on radiological findings in patients accessing the emergency department: a multicentric study.

Purpose: The aim of this multicentric study is to illustrate how the COVID-19 pandemic lockdown affected the workload and outcomes of radiological examinations in emergency radiology. Material and methods: The exams performed in the radiology departments of 4 Italian hospitals during 3 weeks of the Italian lockdown were retrospectively reviewed and compared to the exams conducted during the same period in 2019. Only exams from the emergency department (ED) were included. Two radiologists from each hospital defined the cases as positive or negative findings, based on independent blind readings of the imaging studies. In the case of differences in the evaluation, consensus was reached amongst them via discussion. Continuous measurements are presented as median and interquartile range, while categorical measurements are presented as frequency and percentage; p-values were calculated using the t-test, Mann-Whitney test, and the c2 test. Results: There were 745 patients (53% male; 62 years [44-78]) who underwent radiological examinations in 2020 vs. 2623 (52% male; 56 years [35-76]) in 2019 (p < 0.001). Furthermore, the total number of ED exams dropped from 3206 (2019) to 939 (2020), with a relative increase of CT examinations from 23% to 33% (p < 0.001). The percentage of patients with a positive finding was significantly higher in 2020 (355, 48%) compared to 2019 (684, 26%) (p < 0.001). Conclusions: Our findings show that despite the reduction of emergency radiological examinations, there was a rela-tive increase in the number of positive cases. These significant findings are crucial to ensure better organization of radiology departments and improve patient management during similar health emergencies in the future.

Diagnostic Accuracy and Performance of Artificial Intelligence in Detecting Lung Nodules in Patients With Complex Lung Disease: A Noninferiority Study.

OBJECTIVES: The aim of the study is to investigate the performance of artificial intelligence (AI) convolutional neural networks (CNN) in detecting lung nodules on chest computed tomography of patients with complex lung disease, and demonstrate its noninferiority when compared against an experienced radiologist through clinically relevant assessments. METHODS: A CNN prototype was used to retrospectively evaluate 103 complex lung disease cases and 40 control cases without reported nodules. Computed tomography scans were blindly evaluated by an expert thoracic radiologist; a month after initial analyses, 20 positive cases were re-evaluated with the assistance of AI. For clinically relevant applications: (1) AI was asked to classify each patient into nodules present or absent and (2) AI results were compared against standard radiology reports. Standard statistics were performed to determine detection performance. RESULTS: AI was, on average, 27 seconds faster than the expert and detected 8.4% of nodules that would have been missed. AI had a sensitivity of 67.7%, similar to an accuracy reported for experienced radiologists. AI correctly classified each patient (nodules present/absent) with a sensitivity of 96.1%. When matched against radiology reports, AI performed with a sensitivity of 89.4%. Control group assessment demonstrated an overall specificity of 82.5%. When aided by AI, the expert decreased the average assessment time per case from 2:44 minutes to 35.7 seconds, while reporting an overall increase in confidence. CONCLUSION: In a group of patients with complex lung disease, the sensitivity of AI is similar to an experienced radiologist and the tool helps detect previously missed nodules. AI also helps experts analyze for lung nodules faster and more confidently, a feature that is beneficial to patients and favorable to hospitals due to increased patient load and need for shorter turnaround times.

Evaluating a New Contrast Media Injection System in Coronary CT Angiography.

PURPOSE: To evaluate a new contrast media (CM) injection system in patients undergoing coronary computed tomography angiography (CCTA). METHODS: Seventy-one consecutive patients (33 men and 38 women, mean age 59.0 ± 14.5 years) who underwent CCTA between February and April 2019 using the CT injection system MEDRAD Stellant FLEX (Bayer) were included retrospectively in this single-center study. Quantitative and qualitative image quality parameters were assessed, and the injection system's usability and operational efficiency were evaluated. Results were compared with a matched control group. RESULTS: All examinations were rated as diagnostic. Usability and operational efficiency of the new injector were rated higher than that of the standard injector system, and no significant differences were found for quantitative and qualitative image quality parameters compared with the control group (P ≥ .05). DISCUSSION: Software-based injection facilitates individualized CM application while maintaining high image quality standards in CCTA. Diagnostic accuracy analysis was not performed, but as image quality analysis showed no significant differences, no discrepancies regarding this issue are expected. CONCLUSION: This study demonstrates that the MEDRAD Stellant FLEX CT injection system allows for consistent high-quality CCTA scanning with increased usability and operational efficiency.

Evaluation of a Tube Voltage-Tailored Contrast Medium Injection Protocol for Coronary CT Angiography: Results From the Prospective VOLCANIC Study.

OBJECTIVE. The purpose of this study was to prospectively evaluate, using software support, the feasibility and the quantitative and qualitative image quality parameters of a tube voltage-tailored contrast medium (CM) application protocol for patient-specific injection during coronary CT angiography (CCTA). SUBJECTS AND METHODS. In the Voltage-Based Contrast Media Adaptation in Coronary Computed Tomography Angiography (VOLCANIC-CTA) study, a single-center trial, 120 patients referred for CCTA were prospectively assigned to a tube voltage-tailored CM injection protocol. Automated tube voltage levels were selected in 10-kV intervals and ranged from 70 to 130 kV, and the iodine delivery rate (IDR) was adapted to the tube voltage level using dedicated software. The administered CM volume (370 mg I/mL) ranged from 33 mL at 70 kV (IDR, 0.7 g I/s) to 65 mL at 130 kV (IDR, 1.7 g I/s). Attenuation was measured in the aorta and coronary arteries to calculate quantitative signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR), and 5-point scales were used to evaluate overall image quality. Radiation metrics were also assessed and compared among the protocols. RESULTS. The mean age of the study patients was 62.5 ± 11.9 (SD) years. Image quality was rated as diagnostic in all patients. Contrast attenuation peaked at 70 kV (p < 0.001), whereas SNR and CNR parameters showed no significant differences between tube voltage levels (p ≥ 0.085). Additionally, no significant differences in subjective image quality parameters were found among the different protocols (p ≥ 0.139). The lowest radiation dose values were observed in the group assigned to the 70-kV protocol, which had a median radiation effective dose of 2.0 mSv (p < 0.001). CONCLUSION. The proposed tube voltage-tailored injection protocol allows individualized scanning of patients undergoing CCTA and significantly reduces CM and radiation dose while maintaining a high diagnostic image quality.

Individualized coronary calcium scoring at any tube voltage using a kV-independent reconstruction algorithm.

PURPOSE: We prospectively investigate the feasibility of a patient specific automated tube voltage selection (ATVS)-based coronary artery calcium scoring (CACS) protocol, using a kV-independent reconstruction algorithm, to achieve significant dose reductions while maintaining the overall cardiac risk classification. METHODS: Forty-three patients (mean age, 61.8 ± 9.0 years; 40% male) underwent a clinically indicated CACS scan at 120kVp, as well as an additional CACS acquisition using an individualized tube voltage between 70 and 130kVp based on the ATVS selection (CARE-kV). Datasets of the additional CACS scans were reconstructed using a kV-independent algorithm that allows for calcium scoring without changing the weighting threshold of 130HU, regardless of the tube voltage chosen for image acquisition. Agatston scores and radiation dose derived from the different ATVS-based CACS studies were compared to the standard acquisition at 120kVp. RESULTS: Thirteen patients displayed a score of 0 and were correctly identified with the ATVS protocol. Agatston scores derived from the standard 120kVp (median, 33.4; IQR, 0-289.7) and the patient-tailored kV-independent protocol (median, 47.5; IQR, 0-287.5) showed no significant differences (p = 0.094). The intra-class correlation for Agatston scores derived from the two different protocols was excellent (ICC = 0.99). The mean dose-length-product was 29.8 ± 11.9 mGy × cm using the ATVS protocol and 31.7 ± 11.4 mGy × cm using the standard 120kVp protocol (p < 0.001). Additionally, 95% of patients were classified into the same risk category (0, 1-10, 11-100, 101-400, or > 400) using the patient-tailored protocol. CONCLUSIONS: ATVS-based CACS, using a kV-independent algorithm, allows for high accuracy compared to the standard 120kVp scanning, while significantly reducing radiation dose parameters. KEY POINTS: • ATVS allows for CT scanning with reduced radiation dose values. • KV-independent CACS is feasible at any tube voltage between 70 and 130 kVp. • ATVS applied to kV-independent CACS can significantly reduce the radiation dose.

How the Workload and Outcome of Imaging Examinations Changed During the COVID-19 Pandemic Lockdown.

BACKGROUND: On March 9th, 2020, the Italian government decided to go into lockdown due to the COVID-19 pandemic, which led to changes in the workflow of radiological examinations. AIMS: Aim of the study is to illustrate how the workload and outcome of radiological exams changed in a community hospital during the pandemic. METHODS AND MATERIAL: The exams performed in the radiology department from March 9th to March 29th, 2020 were retrospectively reviewed and compared to the exams conducted during the same time-period in 2019. Only exams coming from the emergency department (ED) were included. Two radiologists defined the cases as positive or negative findings, based on independent blind readings of the imaging studies. Categorical measurements are presented as frequency and percentages, and p-values are calculated using the Chi-squared test. RESULTS AND CONCLUSIONS: There was a significant reduction in the amount of exams performed in 2020: there were 143 (93|65% male, 60.7±21.5 years) patients who underwent radiological examinations from the ED vs. 485 (255|53% male, 51.2±24.8 years) in 2019. Furthermore, the total number of ED exams dropped from 699 (2019) to 215 (2020). However, the percentage of patients with a positive result was significantly higher in 2020 (69|48%) compared to 2019 (151|31%) (p<.001). The reduction of emergency radiological examinations might be a result of the movement restrictions enforced during the lockdown, and possible fear of the hospital as a contagious place. This translated to a relative increase of positive cases as only patients with very serious conditions were accessing the ED.

Low-kV coronary artery calcium scoring with tin filtration using a kV-independent reconstruction algorithm.

PURPOSE: To investigate the accuracy of Agatston scoring and potential for radiation dose reduction of a coronary artery calcium scoring (CACS) CT protocol at 100 kV with tin filtration (Sn100kV) and kV-independent iterative reconstruction, compared to standard 120 kV acquisitions. MATERIALS AND METHODS: With IRB approval and in HIPAA compliance, 114 patients (61.8 ± 9.6 years; 66 men) underwent CACS using a standard 120 kV protocol and an additional Sn100kV CACS scan. The two datasets were reconstructed using a medium sharp convolution algorithm and in addition the Sn100kV scans were reconstructed iteratively based on a kV-independent algorithm. Agatston scores and radiation dose values were compared between the Sn100kV and the standard 120 kV protocol. RESULTS: Median Agatston scores derived from the Sn100kV protocol with the kV-independent algorithm and the standard 120 kV were 21.4 (IQR, 0-173.8) and 24.7 (IQR, 0-171.1) respectively, with no significant differences (p=0.18). Agatston scores derived from the two different protocols had an excellent correlation (r = 0.99). The dose-length-product was 11.5 ± 4.1 mGy × cm using Sn100kV and 50.4 ± 24.9 mGy × cm using the standard 120 kV protocol (p < 0.01), resulting in a significantly lower (77%) effective dose at Sn100kV (0.16 ± 0.06 mSv vs. 0.71 ± 0.35 mSv, p < 0.01). Additionally, 99% of the patients were classified into the same risk category (0, 1-10, 11-100, 101-400, or >400) using the Sn100kV protocol. CONCLUSION: CACS at Sn100kV using the kV-independent iterative algorithm is feasible and provides high accuracy when compared to standard 120 kV scanning. Furthermore, radiation dose can be significantly reduced for this screening application in a priori healthy individuals.

Myocardial extracellular volume fraction to differentiate healthy from cardiomyopathic myocardium using dual-source dual-energy CT.

OBJECTIVE: To evaluate the feasibility of dual-energy CT (DECT)-based iodine quantification to estimate myocardial extracellular volume (ECV) fraction in patients with and without cardiomyopathy (CM), as well as to assess its ability to distinguish healthy myocardial tissue from cardiomyopathic, with the goal of defining a threshold ECV value for disease detection. METHODS: Ten subjects free of heart disease and 60 patients with CM (mean age 66.4 ±â€¯9.4; 59 males and 11 females; 40 ischemic and 20 non-ischemic CM) underwent late iodine enhanced DECT imaging. Myocardial iodine maps were obtained using 3-material decomposition. ECV of the left ventricle was estimated from hematocrit levels and the iodine maps using the AHA 16-segment model. Receiver operating characteristic curve analysis was performed, with corresponding area under the curve, along with Youden's index assessment, to establish a threshold for CM detection. RESULTS: The median ECV for healthy myocardium, non-ischemic CM, and ischemic CM were 25.4% (22.9-27.3), 38.3% (33.7-43.0), and 36.9% (32.4-41.1), respectively. Healthy myocardium showed significantly lower ECV values compared to ischemic and non-ischemic CM (p < 0.001). From Youden's index analysis, an ECV>29.5% would indicate the presence of CM in the myocardium (sensitivity = 90.3; specificity = 90.3); the AUC for this criterion was 0.950 (p < 0.001). CONCLUSION: The findings of this study resulted in a statistically significant distinction between healthy myocardium and CM ECVs. This led to the establishment of a promising threshold ECV value that could facilitate the differentiation between healthy and diseased myocardium, and highlights the potential of this DECT methodology to detect cardiomyopathic tissue.

Cross-modality Accuracy of Dual-step, Prospectively Electrocardiography-triggered Dual-source Computed Tomorgaphy Compared With Same-day Echocardiography and Cardiac Magnetic Resonance Imaging in the Follow-up of Heart-transplant Patients.

PURPOSE: An accurate evaluation of left ventricular volumes, mass, and ejection fraction (EF) and an early exclusion or detection of significant coronary artery disease or cardiac allograft vasculopathy are mandatory for clinical management and prognosis assessment of heart-transplanted patients (HTP). The purpose of this article was to evaluate the role of dual-step prospective electrocardiography-triggered Dual-Source CT (pECGdual-step-DSCT) in HTP for the assessment of left-ventricular function, in comparison with echocardiography (echo) and cardiac magnetic resonance (CMR) performed on the same day, and of the coronary arteries as well. MATERIALS AND METHODS: Left-ventricular EF, end-diastolic volume, end-systolic volume, stroke volume, cardiac output (CO), and mass were assessed in 11 HTP by pECGdual-step-DSCT in comparison with CMR and echo performed on the same day. During all the examinations, the heart rate was recorded. CT coronary artery assessment was also performed. RESULTS: Heart rate was lower during DSCT (75.6±7.8 bpm; P<0.001). EF resulted slightly lower for DSCT (55.7%±5.0%; P≥0.05) in comparison with CMR (57.8%±5.3%; P≥0.05) and echo (59.2%±5.6%; P≥0.05). DSCT showed statistically significant higher end-diastolic volume (153.7±24.2 mL), end-systolic volume (67.8±11.5 mL), and stroke volume (85.9±17.6 mL) (P<0.01 up to 0.001) than CMR, but with a high correlation (P<0.001). Cardiac output was almost similar for DSCT versus CMR, with a very high correlation coefficient (r=0.914; P<0.001). DSCT showed higher mass values than CMR (P<0.001), but with a high correlation (r=0.866; P<0.001). DSCT versus echo results were less correlated. No significant coronary artery disease was detected. CONCLUSION: pECGdual-step-DSCT allows reliable assessment of left-ventricular function in HTP, with good agreement and correlation with CMR, within a global diagnostic approach including coronary artery evaluation in one single-volume acquisition.