Predicting Mechanical Ventilation and Mortality in COVID-19 Using Radiomics and Deep Learning on Chest Radiographs: A Multi-Institutional Study.

In this study, we aimed to predict mechanical ventilation requirement and mortality using computational modeling of chest radiographs (CXRs) for coronavirus disease 2019 (COVID-19) patients. This two-center, retrospective study analyzed 530 deidentified CXRs from 515 COVID-19 patients treated at Stony Brook University Hospital and Newark Beth Israel Medical Center between March and August 2020. Linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), and random forest (RF) machine learning classifiers to predict mechanical ventilation requirement and mortality were trained and evaluated using radiomic features extracted from patients' CXRs. Deep learning (DL) approaches were also explored for the clinical outcome prediction task and a novel radiomic embedding framework was introduced. All results are compared against radiologist grading of CXRs (zone-wise expert severity scores). Radiomic classification models had mean area under the receiver operating characteristic curve (mAUCs) of 0.78 ± 0.05 (sensitivity = 0.72 ± 0.07, specificity = 0.72 ± 0.06) and 0.78 ± 0.06 (sensitivity = 0.70 ± 0.09, specificity = 0.73 ± 0.09), compared with expert scores mAUCs of 0.75 ± 0.02 (sensitivity = 0.67 ± 0.08, specificity = 0.69 ± 0.07) and 0.79 ± 0.05 (sensitivity = 0.69 ± 0.08, specificity = 0.76 ± 0.08) for mechanical ventilation requirement and mortality prediction, respectively. Classifiers using both expert severity scores and radiomic features for mechanical ventilation (mAUC = 0.79 ± 0.04, sensitivity = 0.71 ± 0.06, specificity = 0.71 ± 0.08) and mortality (mAUC = 0.83 ± 0.04, sensitivity = 0.79 ± 0.07, specificity = 0.74 ± 0.09) demonstrated improvement over either artificial intelligence or radiologist interpretation alone. Our results also suggest instances in which the inclusion of radiomic features in DL improves model predictions over DL alone. The models proposed in this study and the prognostic information they provide might aid physician decision making and efficient resource allocation during the COVID-19 pandemic.

Subdural hygroma after spontaneous rupture of an arachnoid cyst in a pediatric patient: A case report.

Arachnoid cysts are benign masses that represent a relatively small percentage of intracranial lesions. Spontaneous rupture of an arachnoid cyst resulting in a subdural hygroma is a very rare event. We report a case of a pediatric patient with a history of an arachnoid cyst and chronic headaches presenting with bilateral papilledema, worsening headaches, and no history of head trauma. Magnetic resonance imaging of the brain revealed an extra-axial cystic lesion in the right middle cranial fossa, similar to an arachnoid cyst seen on previous imaging. A new right subdural collection similar to the cerebral spinal fluid signal causing mass effect on brain parenchyma was determined to represent a subdural hygroma. Craniotomy was performed to evacuate the subdural hygroma as well as cyst fenestration. We report this case to emphasize the importance of considering spontaneous rupture of an arachnoid cyst as a differential diagnosis despite absence of head trauma.

Predicting Clinical Outcomes in COVID-19 using Radiomics and Deep Learning on Chest Radiographs: A Multi-Institutional Study.

We predict mechanical ventilation requirement and mortality using computational modeling of chest radiographs (CXRs) for coronavirus disease 2019 (COVID-19) patients. This two-center, retrospective study analyzed 530 deidentified CXRs from 515 COVID-19 patients treated at Stony Brook University Hospital and Newark Beth Israel Medical Center between March and August 2020. DL and machine learning classifiers to predict mechanical ventilation requirement and mortality were trained and evaluated using patient CXRs. A novel radiomic embedding framework was also explored for outcome prediction. All results are compared against radiologist grading of CXRs (zone-wise expert severity scores). Radiomic and DL classification models had mAUCs of 0.78+/-0.02 and 0.81+/-0.04, compared with expert scores mAUCs of 0.75+/-0.02 and 0.79+/-0.05 for mechanical ventilation requirement and mortality prediction, respectively. Combined classifiers using both radiomics and expert severity scores resulted in mAUCs of 0.79+/-0.04 and 0.83+/-0.04 for each prediction task, demonstrating improvement over either artificial intelligence or radiologist interpretation alone. Our results also suggest instances where inclusion of radiomic features in DL improves model predictions, something that might be explored in other pathologies. The models proposed in this study and the prognostic information they provide might aid physician decision making and resource allocation during the COVID-19 pandemic.

Effects of Prone Ventilation on Oxygenation, Inflammation, and Lung Infiltrates in COVID-19 Related Acute Respiratory Distress Syndrome: A Retrospective Cohort Study.

Patients receiving mechanical ventilation for coronavirus disease 2019 (COVID-19) related, moderate-to-severe acute respiratory distress syndrome (CARDS) have mortality rates between 76-98%. The objective of this retrospective cohort study was to identify differences in prone ventilation effects on oxygenation, pulmonary infiltrates (as observed on chest X-ray (CXR)), and systemic inflammation in CARDS patients by survivorship and to identify baseline characteristics associated with survival after prone ventilation. The study cohort included 23 patients with moderate-to-severe CARDS who received prone ventilation for ≥16 h/day and was segmented by living status: living (n = 6) and deceased (n = 17). Immediately after prone ventilation, PaO2/FiO2 improved by 108% (p < 0.03) for the living and 150% (p < 3 × 10-4) for the deceased. However, the 48 h change in lung infiltrate severity in gravity-dependent lung zones was significantly better for the living than for the deceased (p < 0.02). In CXRs of the lower lungs before prone ventilation, we observed 5 patients with confluent infiltrates bilaterally, 12 patients with ground-glass opacities (GGOs) bilaterally, and 6 patients with mixed infiltrate patterns; 80% of patients with confluent infiltrates were alive vs. 8% of patients with GGOs. In conclusion, our small study indicates that CXRs may offer clinical utility in selecting patients with moderate-to-severe CARDS who will benefit from prone ventilation. Additionally, our study suggests that lung infiltrate severity may be a better indicator of patient disposition after prone ventilation than PaO2/FiO2.

Juvenile idiopathic arthritis in the lumbosacral spine of a toddler: Periarticular osteopenia in a rare location.

Juvenile idiopathic arthritis (JIA) is a group of childhood inflammatory arthropathies that affects multiple joints including the spine, particularly the cervical region. There is paucity of literature regarding JIA in the lumbosacral spine; the few published studies which discuss imaging findings in the lumbosacral spine only include cohorts of older children and adolescents. We present a 22-month-old boy with refusal to walk, in which plain radiographs and contrast-enhanced magnetic resonance imaging of the lumbosacral spine suggested a diagnosis of JIA.

Sister Mary Joseph nodule: an often overlooked or misdiagnosed entity on imaging.

Sister Mary Joseph nodule (SMJN) is an umbilical nodule representing a metastatic deposit from an intra-abdominal primary malignancy. Most radiologists are unaware of this phenomenon, and cases of SMJN have rarely been described in the radiology literature, to our knowledge. We present an example of a patient with known primary pancreatic adenocarcinoma found to have an umbilical nodule as the first manifestation of metastatic disease after an initial misdiagnosis on computed tomography. In addition, we delineate the importance of maintaining a high index of suspicion and pattern recognition for SMJN during imaging when a patient presents with umbilical pain in the setting of known malignancy, since early diagnosis can alter management.

Early Diagnosis of COVID-19 on Non-Chest CT Studies in the Emergency Setting: A Case Series and Review of the Literature.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a novel strain of coronavirus that has spread throughout the globe causing coronavirus disease 2019 (COVID-19). As the number of cases rises in the United States (US), it has become more imperative to detect COVID-19 at its earliest radiologic stage to decrease community transmission. In this case series, we discuss five patients who presented with non-respiratory-related symptoms and underwent non-chest CT imaging, such as abdominal and neck CT, with a portion of the lungs visualized in each respective study. Imaging findings of COVID-19 include basilar and peripherally predominant pulmonary parenchymal ground-glass opacities. All five of our patients had findings suggestive of COVID-19 that prompted the radiologist to suggest testing for the disease. Subsequently, four of the five patients tested positive for COVID-19, and one of them was presumed to have the diagnosis based on clinical and imaging findings.

Vaping-associated acute respiratory distress syndrome.

Vaping-associated lung injury via the use of electronic nicotine delivery systems (ENDS) is currently being evaluated as a potential source of pulmonary injury with uncertain etiology as the use of tetrahydrocannabinol (THC) is increasing throughout the USA. ENDS are marketed to be unlike traditional cigarette smoking in that they are purported to contain only propylene glycol, vegetable glycerine, nicotine, and flavorants compared with the > 60 carcinogenic ingredients in cigarettes. The New England Journal of Medicine (NEJM) currently reports four imaging patterns correlated with vaping-attributed pathology including acute eosinophilic pneumonia, diffuse alveolar damage, organizing pneumonia, and lipoid pneumonia. The incidence and extent of lung disease in otherwise young healthy patients with a history of vaping has not however been definitively recognized within the field of radiology. We present a case of vaping-associated acute respiratory distress syndrome (ARDS) in a young patient with no additional past medical history. The immediate radiologic recognition of vaping as a risk factor for ARDS in the emergency setting is pivotal so that appropriate medical management and respiratory support can be initiated without delay.