Therapy-Induced Evolution of Human Lung Cancer Revealed by Single-Cell RNA Sequencing.

Lung cancer, the leading cause of cancer mortality, exhibits heterogeneity that enables adaptability, limits therapeutic success, and remains incompletely understood. Single-cell RNA sequencing (scRNA-seq) of metastatic lung cancer was performed using 49 clinical biopsies obtained from 30 patients before and during targeted therapy. Over 20,000 cancer and tumor microenvironment (TME) single-cell profiles exposed a rich and dynamic tumor ecosystem. scRNA-seq of cancer cells illuminated targetable oncogenes beyond those detected clinically. Cancer cells surviving therapy as residual disease (RD) expressed an alveolar-regenerative cell signature suggesting a therapy-induced primitive cell-state transition, whereas those present at on-therapy progressive disease (PD) upregulated kynurenine, plasminogen, and gap-junction pathways. Active T-lymphocytes and decreased macrophages were present at RD and immunosuppressive cell states characterized PD. Biological features revealed by scRNA-seq were biomarkers of clinical outcomes in independent cohorts. This study highlights how therapy-induced adaptation of the multi-cellular ecosystem of metastatic cancer shapes clinical outcomes.

Imaging Manifestations of Chest Trauma.

Trauma is the leading cause of death among individuals under 40 years of age, and pulmonary trauma is common in high-impact injuries. Unlike most other organs, the lung is elastic and distensible, with a physiologic capacity to withstand significant changes in contour and volume. The most common types of lung parenchymal injury are contusions, lacerations, and hematomas, each having characteristic imaging appearances. A less common type of lung injury is herniation. Chest radiography is often the first-line imaging modality performed in the assessment of the acutely injured patient, although there are inherent limitations in the use of this modality in trauma. CT images are more accurate for the assessment of the nature and extent of pulmonary injury than the single-view anteroposterior chest radiograph that is typically obtained in the trauma bay. However, the primary limitations of CT concern the need to transport the patient to the CT scanner and a longer processing time. The American Association for the Surgery of Trauma has established the most widely used grading scale to describe lung injury, which serves to communicate severity, guide management, and provide useful prognostic factors in a systematic fashion. The authors provide an in-depth exploration of the most common types of pulmonary parenchymal, pleural, and airway injuries. Injury grading, patient management, and potential complications of pulmonary injury are also discussed. ©RSNA, 2021.

Integrated PET-MRI for Glioma Surveillance: Perfusion-Metabolism Discordance Rate and Association With Molecular Profiling.

OBJECTIVE: Both 18F-FDG PET and perfusion MRI are commonly used techniques for posttreatment glioma surveillance. Using integrated PET-MRI, we assessed the rate of discordance between simultaneously acquired FDG PET images and dynamic contrast-enhanced (DCE) perfusion MR images and determined whether tumor genetics predicts discordance. MATERIALS AND METHODS: Forty-one consecutive patients with high-grade gliomas (20 with grade IV gliomas and 21 with grade III gliomas) underwent a standardized tumor protocol performed using an integrated 3-T PET-MRI scanner. Quantitative measures of standardized uptake value, plasma volume, and permeability were obtained from segmented whole-tumor volumes of interest and targeted ROIs. ROC curve analysis and the Youden index were used to identify optimal cutoffs for FDG PET and DCE-MRI. Two-by-two contingency tables and percent agreement were used to assess accuracy and concordance. Twenty-six patients (63%) from the cohort underwent next-generation sequencing for tumor genetics. RESULTS: The best-performing FDG PET and DCE-MRI cutoffs achieved sensitivities of 94% and 91%, respectively; specificities of 56% and 89%, respectively; and accuracies of 80% and 83%, respectively. FDG PET and DCE-MRI findings were discordant for 11 patients (27%), with DCE-MRI findings correct for six of these patients (55%). Tumor grade, tumor volume, bevacizumab exposure, and time since radiation predicted discordance between FDG PET and DCE-MRI findings, with an ROC AUC value of 0.78. Isocitrate dehydrogenase gene and receptor tyrosine kinase gene pathway mutations increased the ROC AUC value to 0.83. CONCLUSION: FDG PET and DCE-MRI show comparable accuracy and sensitivity in identifying tumor progression. These modalities were shown to have discordant findings for more than a quarter of the patients assessed. Tumor genetics may contribute to perfusion-metabolism discordance, warranting further investigation.

CT-Guided Percutaneous Lung Biopsies in Patients With Suspicion for Infection May Yield Clinically Useful Information.

OBJECTIVE: The objective of our study was to assess the frequency and time frame with which CT-guided lung biopsies for suspected infection yield information that can affect patient management. MATERIALS AND METHODS: All CT-guided lung biopsies over a 68-month period performed for the purpose of diagnosing a suspected infection were reviewed to determine the proportion that yielded information affecting patient management. Patients were included if infection was the only consideration causing the pulmonary lesion in question. RESULTS: Twenty-one biopsies were performed to identify a specific organism causing infection in patients with suspected infection; all patients were receiving antibiotics, 20 (95%) were immunocompromised, and 15 (71%) had undergone a prior bronchoscopy. Material collected from the biopsy provided a diagnosis in nine (43%) patients, whereas the biopsy results were nondiagnostic in the remaining 12 (57%). Of the nine patients for whom the biopsy yielded a diagnosis, eight biopsies revealed the species causing an infection (38%) and one biopsy (5%) detected posttransplant lymphoproliferative disease. Of the nine diagnoses, management was changed as a result of the biopsy in six patients (29% of all patients). The organisms identified by CT-guided lung biopsy in eight patients were fungi of the order Mucorales (i.e., mucormycosis) (n = 3), Aspergillus (n = 3), Pseudomonas (n = 1), and Nocardia (n = 1). The mean elapsed time between biopsy and pathologic diagnosis was 4 days (median, 3 days). There was no association between prior bronchoscopy and nondiagnostic biopsy results. CONCLUSION: CT-guided lung biopsies in patients with a high pretest suspicion for infection result in information sufficient to change patient management in 29% of patients. Organisms identified in these patients were most frequently fungi.

Pediatric chest CT at chest radiograph doses: when is the ultralow-dose chest CT clinically appropriate?

PURPOSE: Computed tomography (CT) use in emergency departments represents a significant contribution to pediatric patients' exposure to ionizing radiation. Here, we evaluate whether ultralow-dose chest CT can be diagnostically adequate for other diagnoses and whether model-based iterative reconstruction (MBIR) can improve diagnostic adequacy compared to adaptive statistical iterative reconstruction (ASIR) at ultralow doses. METHODS: Twenty children underwent chest CTs: 10 standard-dose reconstructed with ASIR and 10 ultralow-dose reconstructed with ASIR and MBIR. Four radiologists assessed images for their adequacy to exclude five hypothetical diagnoses: foreign body, fracture, lung metastasis, pulmonary infection, and interstitial lung disease. Additionally, pairwise comparison for subjective image quality was used to compare ultralow-dose chest CT with ASIR and MBIR. Radiation dose and objective image noise measures were obtained. RESULTS: For exclusion of an airway foreign body, the adequacy of ultralow-dose CT was comparable to standard-dose (p = 0.6). For the remaining diagnoses, ultralow-dose CT was inferior to standard-dose (p = 0.03-<0.001). MBIR partially recovered the adequacy of ultralow-dose CT to exclude pulmonary infection (p = 0.017), but was suboptimal for the other diagnoses. Image noise was significantly lower with MBIR compared to ASIR in ultralow-dose CT (p < 0.001), although subjective preference showed only a slight advantage of MBIR (58 versus 42%). CONCLUSIONS: Ultralow-dose chest CT may be adequate for airway assessment, but suboptimal for the evaluation parenchymal lung disease. Although MBIR improves objective and subjective image quality, it does not completely restore the diagnostic adequacy of ultralow-dose CT when compared to standard-dose CT.

Tests, Quizzes, and Self-Assessments: How to Construct a High-Quality Examination.

OBJECTIVE: The purposes of this article are to highlight aspects of tests that increase or decrease their effectiveness and to provide guidelines for constructing high-quality tests in radiology. CONCLUSION: Many radiologists help construct tests for a variety of purposes. Only well-constructed tests can provide reliable and valuable information about the test taker.

Assessing the learning potential of an interactive digital game versus an interactive-style didactic lecture: the continued importance of didactic teaching in medical student education.

BACKGROUND: Games with educational intent offer a possible advantage of being more interactive and increasing learner satisfaction. OBJECTIVE: We conducted a two-armed experiment to evaluate student satisfaction and content mastery for an introductory pediatric radiology topic, taught by either an interactive digital game or with a traditional didactic lecture. MATERIALS AND METHODS: Medical students participating in a fourth-year radiology elective were invited to participate. Student cohorts were alternatively given a faculty-supervised 1h session playing a simple interactive digital Tic-tac-toe quiz module on pediatric gastrointestinal radiology or a 1h didactic introductory lecture on the same topic. Survey questions assessed the learners' perceived ability to recall the material as well as their satisfaction with the educational experience. Results of an end-of-rotation exam were reviewed to evaluate a quantitative measure of learning between groups. Survey responses were analyzed with a chi-squared test. Exam results for both groups were analyzed with a paired Student's t-test. RESULTS: Students in the lecture group had higher test scores compared to students in the game group (4.0/5 versus 3.6/5, P = 0.045). Students in the lecture group reported greater understanding and recall of the material than students in the game group (P < 0.001 and P = 0.004, respectively). Students in the lecture group perceived the lecture to be more enjoyable and a better use of their time compared to those in the game group (P = 0.04 and P < 0.001, respectively). There was no statistically significant difference between the lecture and game group in ability to maintain interest (P = 0.187). In comparison to pre-survey results, there was a statistically significant decrease in interest for further digital interactive materials reported by students in the game group (P = 0.146). CONCLUSION: Our experience supported the use of a traditional lecture over a digital game module. While these results might be affected by the specific lecture and digital content in any given comparison, a digital module is not always the superior option.

Understanding Cost-Effectiveness Analyses: An Explanation Using Three Different Analyses of Lung Cancer Screening.

OBJECTIVE: Cost-effectiveness analyses (CEAs) contribute to informed decision making, at both the practitioner and societal levels; therefore, understanding CEAs is valuable for radiologists. In light of the recently published National Lung Cancer Screening Trial (NLST) CEA, we aim to explain the terminology, methods, and heterogeneity of CEAs. CONCLUSION: We compared the NLST results to two example lung cancer screening CEAs (which do not rely on NLST data). Both examples assessed screening but reached substantially different conclusions.

Pairwise comparison versus Likert scale for biomedical image assessment.

OBJECTIVE: Biomedical imaging research relies heavily on the subjective and semi-quantitative reader analysis of images. Current methods are limited by interreader variability and fixed upper and lower limits. The purpose of this study was to compare the performance of two assessment methods, pairwise comparison and Likert scale, for improved analysis of biomedical images. MATERIALS AND METHODS: A set of 10 images with varying degrees of image sharpness was created by digitally blurring a normal clinical chest radiograph. Readers assessed the degree of image sharpness using two different methods: pairwise comparison and a 10-point Likert scale. Reader agreement with actual chest radiograph sharpness was calculated for each method by use of the Lin concordance correlation coefficient (CCC). RESULTS: Reader accuracy was highest for pairwise comparison (CCC, 1.0) and ranked Likert (CCC, 0.99) scores and lowest for nonranked Likert scores (CCC, 0.83). Accuracy improved slightly when readers repeated their assessments (CCC, 0.87) or had reference images available (CCC, 0.91). CONCLUSION: Pairwise comparison and ranked Likert scores yield more accurate reader assessments than nonranked Likert scores.

Assessment of Left Ventricular Enlargement at Multidetector Computed Tomography.

PURPOSE: Because left ventricular (LV) enlargement (LVE) is indicative of an array of cardiac pathologies, including cardiomyopathic, ischemic, and valvular heart diseases, it is important to recognize it early in the course of these diseases. The recognition of LVE on nongated contrast-enhanced computed tomography (CT) scans should be facilitated by the availability of a dimensional index. To our knowledge, no CT index of LVE has been proposed. Therefore, the study aimed to define whether the maximum LV diameter (LVd) measured on nongated multidetector computed tomography can identify LVE when referencing echocardiography as the diagnostic standard. MATERIALS AND METHODS: The patient population consisted of 438 consecutive patients who had a contrast-enhanced, nongated 16- or 64-detector CT of the chest for evaluation of pulmonary embolism or aortic dissection between January 2006 and March 2008. One hundred fifty-five patients in this group also had an echocardiogram within 2 months of the CT study. The maximum LV cavity size, septal to lateral wall dimension, was measured perpendicularly to the long axis of the left ventricle on the axial CT scans by 2 observers blinded to the echocardiography data.An receiver operating characteristic analysis was performed to identify a highly specific cutoff for the diagnosis of LVE on CT, using echocardiogram as the standard of reference. Interobserver agreement was assessed using Bland-Altman analysis. RESULTS: A total of 84 females and 71 males were evaluated (female to male ratio of 1.09). The mean age for the 155 patients was 58 years. Six percent of these patients had a diagnosis of LVE on 2-dimensional echocardiography. The mean (SEM) LV internal diameter at nongated multidetector computed tomography between the group with normal LV and the group with LVE by echocardiography was 4.4 (0.7) cm for the normal LV and 5.9 (1.2) cm for the dilated LVs (P < 0.0001). With the use of threshold value of LVd of 5.6 cm, a sensitivity of 78%, specificity of 100%, positive likelihood ratio of 113.5, and negative likelihood ratio of 0.22 were calculated. The LVd measurements had an excellent agreement between observers on the Bland-Altman analysis. CONCLUSIONS: Left ventricular enlargement can be reliably identified on nongated contrast-enhanced multidetector CT when the maximum luminal diameter of the LV is greater than 5.6 cm. Nongated contrast-enhanced CT scan can be used to recognize LVE.

The inferior emissary vein: a reliable landmark for right adrenal vein sampling.

BACKGROUND: Right adrenal vein (RAV) catheterization can be a very challenging step in adrenal venous sampling (AVS). Visualization of the inferior emissary vein (IEV) may be an indication of successful RAV catheterization. PURPOSE: To compare the rate of successful RAV sampling in the presence of the IEV. MATERIAL AND METHODS: Retrospective review of all consecutive patients with PA who underwent AVS between April 2009 and April 2012 was performed. A total of 30 patients were identified. Procedural images, cortisol, and aldosterone values obtained from sampling of the RAV and inferior vena cava (IVC) were reviewed. Cortisol measurements obtained from RAV samples were divided by measurements from the infra-renal IVC blood samples in order to calculate the selectivity index (SI). An SI >3 was considered indicative of technically successful RAV sampling. RESULTS: RAV sampling was considered technically successful in 29 out of 30 cases (97%). In cases of successful RAV sampling (29 patients), the IEV was identified in 25 patients (86%). The IEV was visualized in isolation in 16 patients (64%), and in conjunction with visualization of the RAV or right adrenal gland stain in nine patients (36%). The IEV was not visualized in the one case of unsuccessful RAV sampling. Visualizing the IEV had a sensitivity of 86.2% for successful RAV sampling. CONCLUSION: The IEV may serve as a reliable landmark for the RAV during RAV sampling.

Hepatotoxicity after transarterial chemoembolization and transjugular intrahepatic portosystemic shunt: do two rights make a wrong?

PURPOSE: To compare the rates of hepatotoxicity after transarterial chemoembolization for hepatocellular carcinoma (HCC) in patients with and without a transjugular intrahepatic portosystemic shunt (TIPS) who were stratified into comparable risk groups. MATERIALS AND METHODS: A retrospective review of patients with HCC who were treated with transarterial chemoembolization between January 2005 and December 2009 was performed. Of 158 patients with comparable model for end-stage liver disease (MELD) scores, 10 had a patent TIPS. Hepatobiliary severe adverse events (SAEs) occurring after transarterial chemoembolization were documented. In addition, 1-year survival and liver transplantation rate after transarterial chemoembolization were calculated in each group. RESULTS: The incidence of hepatobiliary SAEs after transarterial chemoembolization was nearly two times higher in patients with a TIPS (70%) than in patients without a TIPS (36%; P=.046). The liver transplantation rate 1 year after transarterial chemoembolization was 2.5 times higher in patients with a TIPS (80%) than in patients without a TIPS (32%; P=.004). There was no significant difference in 1-year survival between the two groups after transarterial chemoembolization. CONCLUSIONS: Patients with HCC and a patent TIPS are more likely to develop significant hepatotoxicity after transarterial chemoembolization than comparable patients without a TIPS in place.

Correctly using sensitivity, specificity, and predictive values in clinical practice: how to avoid three common pitfalls.

OBJECTIVE: Radiology is the specialty of imaging-based diagnostic tests. Understanding the science behind evaluating diagnostic test performance is essential for radiologists because we provide care to patients and interact with our colleagues. CONCLUSION: Here, we review the key terminology used and common pitfalls encountered in the literature and in day-to-day discussions of diagnostic test performance.

Cardiovascular MR imaging after surgical correction of tetralogy of Fallot: approach based on understanding of surgical procedures.

Tetralogy of Fallot (TOF) is one of the most common congenital heart diseases for which patients are referred for postoperative magnetic resonance (MR) imaging evaluation. The most common surgical procedures for TOF repair include infundibulectomy, transannular pulmonary artery patch repair, and right ventricle-pulmonary artery conduit placement. In the past few decades, surgery has proved successful, but most patients require repeat imaging throughout their lives. MR imaging is now frequently used for morphologic and functional evaluation after TOF repair. The most common late postoperative sequelae and residual lesions include right ventricular outflow tract aneurysm and dyskinesis, conduit failure, pulmonary regurgitation, tricuspid regurgitation, right ventricular failure, residual main and branch pulmonary artery stenosis, branch pulmonary artery aneurysm, left pulmonary artery kinking, and residual or recurrent ventricular septal defect. The imaging approach for the evaluation of patients with repaired TOF should be guided by the surgical procedure used and the complications that are expected. Knowledge of the most common postoperative problems and their cardiovascular MR imaging appearances is essential for good radiology practice in this clinical setting.

CMR assessment of right ventricular function in patients with combined pulmonary stenosis and insufficiency after correction of tetralogy of Fallot.

BACKGROUND: Tetralogy of Fallot (TOF) is one of the most common types of congenital heart disease and requires prompt surgical correction. Post-correction pulmonary insufficiency (PI) often ensues in adulthood. At times, the PI is accompanied by residual pulmonary stenosis (PS). Little is known regarding right ventricular (RV) function in the setting of combined PS and PI. PURPOSE: To compare cardiac magnetic resonance (CMR) parameters for the assessment of RV function between patients with combined pulmonary stenosis and pulmonary insufficiency (PSPI) and isolated PI following surgical repair of TOF. MATERIAL AND METHODS: Retrospective review of patients with comparable corrected TOF and similar PI was performed. Seventeen patients (median age, 24 years; range, 10-52 years) had combined PSPI and 30 patients (median age, 30 years; range, 6-70 years) had isolated PI. Cine magnetic resonance (MR) images (Philips Medical Systems, Best, The Netherlands) in the short-axis plane were used to calculate end-systolic, end-diastolic, and stroke volumes (RVESV, RVEDV, RVSV) and to measure RV wall thickness. Velocity-encoded cine MR images were used to measure pulmonary regurgitation fraction (PRF) by calculating the ratio of backward flow and total forward flow, obtained from the main pulmonary flow analysis. Peak pressure gradient across the pulmonary valve was obtained from spectral Doppler echocardiography. RESULTS: RVEF was 51 ± 8% in the PSPI patients and 39 ± 11%, in the patients with isolated PI (P = 0.001). Additionally, RV wall thickness was 5.2 ± 0.8 mm in the PSPI patients compared to 2.6 ± 0.9 mm in the isolated PI patients (P = 0.001). RVESVi and RVEDVi were significantly lower (P < 0.05) in patients with combined PSPI (60 ± 21 mL/m(2), 121 ± 35 mL/m(2), respectively) compared to the patients with isolated PI (95 ± 48 mL/m(2), 152 ± 61 mL/m(2), respectively). CONCLUSION: RV function is preserved in patients with PSPI when compared to patients with PI following surgical repair of TOF.

A "High-Reliability Organization" Approach to Improve Trauma Imaging Performance.

PURPOSE: CT is the gold standard for triaging critically ill patients, including in trauma, and its use has increased over time. CT turnaround times (TATs) are frequently targeted for improvement. As opposed to linear reductionist processes such as Lean and Six Sigma, a high-reliability organization (HRO) approach focuses on culture and teams to enable rapid problem solving. The authors evaluated the HRO model to rapidly generate, trial, select, and implement improvement interventions to improve trauma patient CT performance. METHODS: All trauma patients presenting to a single institution's emergency department during a 5-month period were included. Project periods included 2-month preintervention, 1-month wash-in, and 2-month postintervention. Each initial trauma CT encounter during the wash-in and postintervention periods created job briefs in which the radiologist ensured that all involved had the pertinent clinical information and agreed on the imaging needed, thereby creating a shared mental model as well as an opportunity to raise concerns and provide ideas for improvement. RESULTS: A total of 447 patients were included: 145 preintervention, 68 wash-in, and 234 postintervention. The seven selected interventions were trauma text alert; scripted CT technologist-radiologist communication; modification of CT acquisition, processing, sending, and interpretation; and trauma mobile phones. The seven selected interventions reduced trauma patient CT median TATs by 60% (78 vs 31 min, P < .001), demonstrating the effectiveness of an HRO approach to improvement. CONCLUSIONS: An HRO-based approach was rapid in generating, trialing, selecting, and implementing improvement interventions, and the interventions were effective at substantially decreasing trauma patient CT TATs.

Mammography Screening Outreach Through Non-Primary Care-Based Services.

OBJECTIVE: To estimate the proportion of patients visiting urgent care centers or emergency departments or being hospitalized who were not up to date with recommended mammography screening to assess the potential impact of non-primary care-based cancer screening interventions. METHODS: Adult participants from the 2019 National Health Interview Survey were included. Among participants not up to date with breast cancer screening guidelines based on ACR recommendations, the proportion of patients reporting an urgent care, emergency department visit, or hospitalization within the last year was estimated accounting for complex survey sampling design features. Multiple variable logistic regression analyses were then conducted to evaluate the association between sociodemographic characteristics and mammography screening adherence. RESULTS: The study included 9,139 women between the ages of 40 and 74 years without history of breast cancer. Of these respondents, 44.9% did not report mammography screening within the last year. Among participants who did not report mammography screening, 29.2% reported visiting an urgent care center, 21.8% reported visiting an emergency room, and 9.6% reported being hospitalized within the last year. The majority of patients receiving non-primary care-based services, who were not up to date with mammography screening, were from historically underserved groups including Black and Hispanic patients. CONCLUSION: Nearly 10% to 30% of participants who have not obtained recommended breast cancer screening have visited non-primary care-based services including urgent care centers or emergency rooms or have been hospitalized within the last year.

Embedding 3D radiology models in portable document format.

OBJECTIVE: The purpose of this article is to discuss how to convert cross-sectional images into a 3D model and embed them in a Portable Document Format (PDF) file. Four programs are used: OsiriX, MeshLab, Microsoft PowerPoint, and Adobe Acrobat. Step-by-step instructions are provided. CONCLUSION: Embedding 3D radiology models into PDF files is a powerful tool that may be used for clinical, educational, and research purposes.

Radiology Imaging Volume Changes During Discrete COVID-19 Pandemic Waves: Implications for the Delta Variant of Coronavirus and Future Pandemics.

PURPOSE: The aim of this study was to evaluate radiology imaging volumes at distinct time periods throughout the coronavirus disease 2019 (COVID-19) pandemic as a function of regional COVID-19 hospitalizations. METHODS: Radiology imaging volumes and statewide COVID-19 hospitalizations were collected, and four 28-day time periods throughout the COVID-19 pandemic of 2020 were analyzed: pre-COVID-19 in January, the "first wave" of COVID-19 hospitalizations in April, the "recovery" time period in the summer of 2020 with a relative nadir of COVID-19 hospitalizations, and the "third wave" of COVID-19 hospitalizations in November. Imaging studies were categorized as inpatient, outpatient, or emergency department on the basis of patient location at the time of acquisition. A Mann-Whitney U test was performed to compare daily imaging volumes during each discrete 28-day time period. RESULTS: Imaging volumes overall during the first wave of COVID-19 infections were 55% (11,098/20,011; P < .001) of pre-COVID-19 imaging volumes. Overall imaging volumes returned during the recovery time period to 99% (19,915/20,011; P = .725), and third-wave imaging volumes compared with the pre-COVID-19 period were significantly lower in the emergency department at 88.8% (7,951/8,955; P < .001), significantly higher for outpatients at 115.7% (8,818/7,621; P = .008), not significantly different for inpatients at 106% (3,650/3,435; P = .053), and overall unchanged when aggregated together at 102% (20,419/20,011; P = .629). CONCLUSIONS: Medical imaging rebounded after the first wave of COVID-19 hospitalizations, with relative stability of utilization over the ensuing phases of the pandemic. As widespread COVID-19 vaccination continues to occur, future surges in COVID-19 hospitalizations will likely have a negligible impact on imaging utilization.