Purpose The purpose of our research is to evaluate the usefulness of chest X-ray for triaging patients with suspected COVID-19 infection. Methods IRB approval was obtained to allow a retrospective review of adult patients who presented to the Emergency Department with a complaint of fever, cough, dyspnea or hypoxia and had a chest X-ray between 12 March 2020 and 26 March 2020. The initial chest X-ray was graded on a scale of 0-3 with grade 0 representing no alveolar opacities, grade 1: < 1/3 alveolar opacities of the lung, Grade 2: 1/3 to 2/3 lung with alveolar opacities and grade 3: > 2/3 alveolar opacities of the lung. Past medical history of diabetes and hypertension, initial oxygen saturation, COVID-19 testing results, intubation, and outcome were also collected. Results Four hundred ten patient chest X-rays were reviewed. Oxygen saturation and X-ray grade were both significantly associated with the length of stay in hospital, the hazard ratio (HR) of discharge was 1.05 (95% CI [1.01, 1.09], p = 0.017) and 0.61 (95% CI [0.51, 0.73], p < 0.001), respectively. In addition, oxygen saturation and X-ray grade were significant predictors of intubation (odds ratio (OR) of intubation is 0.88 (95% CI [0.81, 0.96], p = 0.004) and 3.69 (95% CI [2.25, 6.07], p < 0.001). Conclusions Initial chest X-ray is a useful tool for triaging those subjects who might have poor outcomes with suspected COVID-19 infection and benefit most from hospitalization.

Pulmonary embolism (PE) is a leading cause of morbidity and mortality worldwide. PE is a complex disease with a highly variable presentation and the available treatment options for PE are expanding rapidly. Anticoagulation (AC), systemic lysis, surgery, and catheter-directed thrombolysis (CDT) play important roles in treating patients with PE. Thus, a multidisciplinary approach to diagnosis, risk stratification, and therapy is required to determine which treatment option is best for a given patient with this complex disease.

Background and Purpose The COVID-19 pandemic acutely disrupted all facets of healthcare, with future implications that are expected to resonate for many years. We investigated the effect of the pandemic on neuroimaging volume, hypothesizing that all representative studies would experience a reduction in volume, with those typically performed in the inpatient setting (noncontrast enhanced CT head and CTA head/neck) taking longer to recover to pre-pandemic volumes compared to studies typically performed in the outpatient setting (MR brain with and without and MR lumbar spine without). Materials and Methods We retrospectively queried our institution’s radiology reporting system to collect weekly data for 1 year following the World Health Organization declaration of a pandemic (11 March 2020–9 March 2021) and compared them to imaging volumes from the previous year (11 March 2019–9 March 2020). We subsequently analyzed quarterly data (e.g., first quarter comparison: 3/11/2020–6/9/2020 was compared to 3/11/2019–6/9/2019). Results All studies experienced decreased volume during the first quarter of the year following onset of the COVID-19 pandemic, with noncontrast enhanced CT head failing to recover to pre-pandemic volumes. CTA head/neck actually surpassed pre-pandemic volume by the second quarter of the year. MRI brain w/wo and MRI lumbar spine without recovered to baseline volume by the second quarter. Conclusion Noncontrast enhanced CT head did not recover pre-pandemic imaging volume. CTA head/neck volume initially decreased, however volume increased above pre-pandemic levels during the second quarter; this finding may be attributable to a prothrombotic state in COVID-19 patients.

Inferior vena cava (IVC) filters are commonly encountered on radiographs of the abdomen, lumbar spine, and pelvis but are typically not identified in study reports according to their type and whether they are retrievable or not. We hypothesized that standard machine learning techniques including image analysis by convolutional neural networks (CNNs) would be able to differentiate between various types of IVC filters, and that these automatic classification methods could serve as a useful method to enrich registries of patients with IVC filters. Such patients may in turn be candidates for IVC filter retrieval. Materials: 530 radiographs of twelve categories of IVC filters were collected and regions of interest labelled manually with type of IVC filter. Standard image augmentation strategies were used to increase the number of training images. Transfer learning techniques were employed to retrain the Inception v3 convolutional neural network (CNN), part of the open source Tensorflow framework, to detect and classify the twelve categories of IVC filters. The accuracy of the retrained network was then evaluated by a set of 24 IVC filter radiographs not previously seen by the network. Results: The retrained network required 14 minutes to train on the augmented set of 5300 radiographs using CPU-only techniques and 8000 learning steps. Final test accuracy achieved was 80%, with training set accuracy up to 96%. In a set of 24 test images that the network had not previously seen, it correctly identified 75% of 12 classes of IVC filters. Confidence scores for correct identification of the filters were mainly clustered between 90-99% (range, 51-99%). The network was inaccurate on images of low brightness and contrast. The network could distinguish between similarappearing filters with high confidence, such as between OptEase and TrapEase filters, and between Tulip and Celect filters. Conclusions: In summary, we have trained a convolutional neural network to accurately identify and classify 12 classes of IVC filters. We anticipate that this will automate the detection and reporting of patients with IVC filters for purposes of registries and filter retrieval practices.

No abstract