Bizarre parosteal osteocondromatous proliferation (BPOP) of the acromion with soft tissue recurrence.

Bizarre parosteal osteochondromatous proliferation (BPOP) is a benign but rare periosteal-originating chondrogenic tumor. It commonly arises from the hands and feet. It is slow-growing and often presents as a painless lump. On imaging, the mass is well-marginated and almost always remains contiguous with the cortical bone. Histologically, the lesion is composed of a disorganized admixture of fibrous tissue, bone, and cartilage with bizarre features. Treatment is surgical and local recurrence is common contiguous with bone. This case report demonstrates an uncommon acromial BPOP with the first reported recurrence not contiguous with the underlying cortex.

Primary intradural extraosseous Ewing's sarcoma of the L3 nerve root: illustrative case.

BACKGROUND: Ewing's sarcoma is an uncommon, aggressive malignancy that typically presents as an osseous lesion, most commonly in children and adolescents. Very rarely Ewing's sarcoma can present as an intradural extramedullary mass mimicking more common tumors. OBSERVATIONS: A 32-year-old female had a left L3 nerve root-associated lesion identified in the setting of recent-onset radiculopathy. Contrast-enhanced magnetic resonance imaging of the lumbar spine was favored to demonstrate a schwannoma or neurofibroma. Hemilaminectomy, facetectomy, and resection of the mass led to improved radiculopathy and a tissue diagnosis of Ewing's sarcoma. Immediate referral to medical oncology facilitated expeditious initiation of adjuvant chemotherapy and radiation. LESSONS: The differential diagnosis for newly identified nerve root-associated tumors should remain broad, including common benign pathologies and rare malignant entities. Tissue remains the gold standard for diagnosis, as preoperative imaging suggested a nerve sheath tumor. Malignant pathologies such as Ewing's sarcoma must be considered, especially in the setting of rapidly progressive symptoms or interval growth on serial imaging. Early diagnosis allows for the timely initiation of comprehensive oncological care. Long-term multidisciplinary follow-up is necessary for the surveillance of disease progression.

Intra-Abdominal and Retroperitoneal Benign Lipomatous Tumors-An Extremely Rare Mimic of Liposarcoma and its Diagnostic Challenge.

Background. Lipomas are common superficial soft tissue tumors of mature adipocytes. In contrast, well-differentiated/dedifferentiated liposarcoma typically presents in the retroperitoneum as large masses. We provide clinicopathologic and follow-up details of 9 retroperitoneal/intra-abdominal benign lipomatous tumors (BLT) and discuss the utility of ancillary fluorescence in situ hybridization (FISH) in distinguishing from their malignant counterparts. Design. Clinicopathologic details and histology of 9 intra-abdominal and retroperitoneal lipomas were studied along with ancillary CD10 immunohistochemistry (IHC) and FISH for MDM2 and CDK4 amplification. Results. There were 6 females and 3 males. Median age at diagnosis was 52 years (range 36-81 years). Seven were identified incidentally and 2 presented with primary complaints. On imaging, 7 were considered suspicious for liposarcoma. Grossly, the tumors ranged from 3.4 to 41.2 cm (median 16.5 cm). Histologically, all cases showed well-differentiated BLT, further classified as lipoma (n = 7; 1 with metaplastic ossification, 2 with prominent vessels, and 4 ordinary lipomas) and lipoma-like hibernoma (n = 2)-the latter 2 showed intramuscular lesions with interspersed brown fat. CD10 IHC showed strong staining in the 2 hibernomas, whereas the staining was weak in the remaining. MDM2 and CDK4 amplification were negative by FISH in all. Follow-up (median 18 months) did not show recurrence on clinical or imaging evaluation. Conclusion. Retroperitoneal/intra-abdominal BLT are extremely rare and are indistinguishable clinically and radiographically from liposarcoma. This necessitates molecular confirmation even when the histology is convincingly benign, for a confident diagnosis. Our cohort shows that conservative excision without removal of abutted organs is sufficient in most cases.

Islet-1 Is Differentially Expressed Among Neuroendocrine and Non-Neuroendocrine Tumors and Its Potential Diagnostic Implication.

Islet-1 (ISL1) plays key roles in programming the epigenome and facilitating the recruitment of additional regulatory factors. Although it has been used as a marker for pancreatic neuroendocrine tumors (PanNETs), ISL1 reactivity in other tumor types are critically missing. ISL1 immunohistochemistry was performed on 147 neuroendocrine tumors (NET) originated in pancreas, gastrointestinal tract, lung, thyroid, parathyroid, pituitary, adrenal medulla, head/neck, genitourinary tract, and skin; and 110 non-neuroendocrine tumors originated in the pancreas, thymus, lung, thyroid, mesothelium, adrenal cortex, stomach, breast, head/neck, skin, and kidney. ISL1 nuclear staining was observed in normal thymic epithelium, pancreatic islets, adrenal medulla, and pituitary gland cells as well as frequently in tumors of these origins: pancreatic NET (78%), paraganglioma/pheochromocytoma (100%), thymoma (82%), and pituitary NET (50%). ISL1 was also variably expressed in certain non-pancreatic NET such as Merkel cell carcinoma (100%), medullary carcinoma of the thyroid (100%), head/neck NEC (80%), genitourinary NEC (71%), lung small cell carcinoma (46%), lung carcinoids (17%), lower intestinal tract NET (93%) but not in upper gastrointestinal tract NET nor parathyroid adenoma. For other non-NETs, focal ISL1 expression was less frequently detected in gastric adenocarcinoma (40%), mesothelioma (29%), adrenal cortical carcinoma (17%), and squamous carcinoma (24%), but not in others tested. ISL1 is not a pan-NE marker as it is consistently lacking in upper gastrointestinal NET and parathyroid adenoma. It is also differentially expressed in thymoma. ISL1 immunohistochemnistry could help to differentiate PanNET and lower intestinal NET from upper gastrointestinal NET and be used as a marker for thymoma.

Myofibroblastoma in the Liver: A Case Report and Review of Literature.

Myofibroblastoma is a rare benign mesenchymal tumor first described in the breast. It is also known as mammary-type myofibroblastoma outside of the breast, more frequently located along the embryonic milk line. Exceptionally, myofibroblastoma can occur at visceral locations. We present a case of myofibroblastoma detected incidentally in the liver. A well-circumscribed mass, grossly measuring 6.2 cm in the liver parenchyma, was found on imaging studies. Histologically, the lesion is characterized by benign spindle cells in a hyalinized collagenous stroma, with positive staining for SMA and ER, focal positivity for CD34, negative for desmin, and loss of RB1. This rare tumor at such an unusual location makes it diagnostically challenging, especially on core biopsy of the lesion. To our knowledge, this is the second case of myofibroblastoma in the liver reported in the English literature and the first such case with a detailed pathology description.

Hepatic Macrophage Types Cluster with Disease Etiology in Chronic Liver Disease and Differ Compared to Normal Liver: Implications for Their Biologic and Diagnostic Role.

Introduction. Macrophages are phenotypically heterogeneous cells that play a vital role in hepatic fibrogenesis. We aimed to compare the macrophage profiles between normal livers and those with various chronic liver diseases in the precirrhotic fibrosis stage. Methods. Immunohistochemistry was performed for three macrophage markers (CD163, CD68, and IBA1) on 48 liver biopsies. Digital image analysis and automated cell count were used to calculate the densities of immunostained cells in two selected regions of interest: the periportal region and the perivenous region. Results. The absolute and relative densities of the macrophage phenotypes in relationship with zones and etiologies showed four distinct patterns by hierarchical cluster analysis: (1) no significant increase in the macrophage densities in either periportal or perivenous regions - nonalcoholic steatohepatitis; (2) significant increase in the selected macrophage densities in both periportal and perivenous regions - Hepatitis C; (3) significant increase in the macrophage densities only in periportal region - alcoholic liver disease, primary sclerosing cholangitis, and primary biliary cholangitis; and (4) significant increase in the densities of all types of macrophages in both periportal and perivenous regions - autoimmune hepatitis. Conclusions. There are distinct macrophage phenotypic and zonal geographic signatures correlating to etiologies of chronic liver disease in the precirrhotic stage.

Molecular identification of an ETV6-RET fusion in a secretory carcinoma associated with a pleomorphic adenoma.

Pleomorphic adenoma (PA) is the most common neoplasm of the salivary glands. Although several carcinomas have been reported to arise from PA, only 1 case of salivary gland secretory carcinoma (SC) ex pleomorphic adenoma has been previously reported. SC is a newly described salivary gland tumor harboring an ETV6-NTRK3 translocation, which is classically observed in secretory carcinoma of the breast, although other translocations have recently been observed. We report the first case of the molecular identification of a rare ETV6-RET translocation in an SC arising from a PA in the submandibular salivary gland (SC ex PA). Our results add to the diversity of tumors that are associated with PA and contribute to the molecular characterization of SC, which will have implications on its diagnosis, prognosis, and treatment.

Case Report: Multilevel Ossification of the Ligamentum Flavum in a Patient With Spinal Osteoblastoma.

Introduction: Spinal osteoblastomas are primary benign bone tumors most commonly presenting as diffuse back pain in young adults. Rarely, spinal osteoblastoma is associated with ossification of the ligamentum flavum (OLF), a form of ectopic bone formation, which can present with myelopathy. This report highlights a unique case of a patient with spinal osteoblastoma, associated OLF, and thoracic myelopathy. Case Description: The patient presented with subtle myelopathy consisting of mid-thoracic back pain, paresthesias, and gait instability. Imaging findings were suggestive of spinal osteoblastoma with multifocal OLF. The patient was consented for thoracic decompression and stabilization at the T6-10 levels. Histopathology confirmed osteoblastoma with associated OLF. At follow up, the patient's neurological symptoms had completely resolved. Conclusion: This case describes management for a rare presentation of osteoblastoma with associated OLF and myelopathy. Surgeons should be wary of disproportionate neurological compromise when spinal osteoblastoma is associated with OLF. Further study is required to elucidate the pathogenesis of this condition.

Artificial intelligence for prediction of COVID-19 progression using CT imaging and clinical data.

OBJECTIVES: Early recognition of coronavirus disease 2019 (COVID-19) severity can guide patient management. However, it is challenging to predict when COVID-19 patients will progress to critical illness. This study aimed to develop an artificial intelligence system to predict future deterioration to critical illness in COVID-19 patients. METHODS: An artificial intelligence (AI) system in a time-to-event analysis framework was developed to integrate chest CT and clinical data for risk prediction of future deterioration to critical illness in patients with COVID-19. RESULTS: A multi-institutional international cohort of 1,051 patients with RT-PCR confirmed COVID-19 and chest CT was included in this study. Of them, 282 patients developed critical illness, which was defined as requiring ICU admission and/or mechanical ventilation and/or reaching death during their hospital stay. The AI system achieved a C-index of 0.80 for predicting individual COVID-19 patients' to critical illness. The AI system successfully stratified the patients into high-risk and low-risk groups with distinct progression risks (p < 0.0001). CONCLUSIONS: Using CT imaging and clinical data, the AI system successfully predicted time to critical illness for individual patients and identified patients with high risk. AI has the potential to accurately triage patients and facilitate personalized treatment. KEY POINT: • AI system can predict time to critical illness for patients with COVID-19 by using CT imaging and clinical data.

Neoadjuvant radiation for cutaneous and soft tissue angiosarcoma.

BACKGROUND AND OBJECTIVES: Neoadjuvant radiation (NRT) is frequently utilized in soft tissue sarcomas to increase local control. Its utility in cutaneous and soft tissue angiosarcoma remains poorly defined. METHODS: This retrospective cohort study was performed using the National Cancer Database (2004-2016) evaluating patients with clinically localized, surgically resected angiosarcomas. Factors associated with receipt of NRT in the overall cohort and margin positivity in treatment naïve patients were identified by univariate and multivariable logistic regression analyses. Survival was assessed using Kaplan-Meier analysis. RESULTS: Of 597 patients, 27 (4.5%) received NRT. Increasing age (odds ratio [OR] 0.95, p = 0.025), tumor size more than or equal to 5 cm (OR 3.16, p = 0.02), and extremity tumor location (OR 3.99, p = 0.04) were associated with receipt of NRT. All patients who received NRT achieved an R0 resection (p = 0.03) compared with 17.9% of patients without NRT. Factors associated with risk of margin positivity included tumor size more than or equal to 5 cm (OR 1.85, p = 0.01), and head/neck location (OR 2.24, p = 0.006). NRT was not significantly associated with improved survival (p = 0.21). CONCLUSIONS: NRT improves rates of R0 resection but is infrequently utilized in cutaneous and soft tissue angiosarcoma. Increased usage of NRT, particularly for patients with lesions more than or equal to 5 cm, or head and neck location, may help achieve complete resections.

Increased tumor-infiltrating lymphocyte density is associated with favorable outcomes in a comparative study of canine histiocytic sarcoma.

Histiocytic sarcoma (HS) is a rare and aggressive tumor in humans with no universally agreed standard of care therapy. Spontaneous canine HS exhibits increased prevalence in specific breeds, shares key genetic and biologic similarities with the human disease, and occurs in an immunocompetent setting. Previous data allude to the immunogenicity of this disease in both species, highlighting the potential for their successful treatment with immunotherapy. Quantification of CD3 tumor-infiltrating lymphocytes (TIL) in five cases of human HS revealed variable intra-tumoral T cell infiltration. Due to the paucity of human cases and lack of current model systems in which to appraise associations between anti-tumor immunity and treatment-outcome in HS, we analyzed clinical data and quantified TIL in 18 dogs that were previously diagnosed with localized HS and treated with curative-intent tumor resection with or without adjuvant chemotherapy. As in humans, assessment of TIL in biopsy tissues taken at diagnosis reveal a spectrum of immunologically "cold" to "hot" tumors. Importantly, we show that increased CD3 and granzyme B TIL are positively associated with favorable outcomes in dogs following surgical resection. NanoString transcriptional analyses revealed increased T cell and antigen presentation transcripts associated with prolonged survival in canine pulmonary HS and a decreased tumor immunogenicity profile associated with shorter survivals in splenic HS. Based on these findings, we propose that spontaneous canine HS is an accessible and powerful novel model to study tumor immunology and will provide a unique platform to preclinically appraise the efficacy and tolerability of anti-cancer immunotherapies for HS.

TERT gene rearrangement in chordomas and comparison to other TERT-rearranged solid tumors.

Chordomas are rare, slow-growing neoplasms thought to arise from the foetal notochord remnant. A limited number of studies that examined the mutational profiles in chordomas identified potential driver mutations, including duplication in the TBXT gene (encoding brachyury), mutations in the PI3K/AKT signaling pathway, and loss of the CDKN2A gene. Most chordomas remain without clear driver mutations, and no fusion genes have been identified thus far. We discovered a novel TERT in-frame fusion involving RPH3AL (exon 5) and TERT (exon 2) in the index chordoma case. We screened a discovery cohort of 18 additional chordoma cases for TERT gene rearrangement by FISH, in which TERT rearrangement was identified in one additional case. In our independent, validation cohort of 36 chordomas, no TERT rearrangement was observed by FISH. Immunohistochemistry optimized for nuclear TERT expression showed at least focal TERT expression in 40/55 (72.7%) chordomas. Selected cases underwent molecular genetic profiling, which showed low tumor mutational burdens (TMBs) without obvious driver oncogenic mutations. We next examined a cohort of 1,913 solid tumor patients for TERT rearrangements, and TERT fusions involving exon 2 were observed in 7/1,913 (0.4%) cases. The seven tumors comprised five glial tumors, and two poorly differentiated carcinomas. In contrast to chordomas, the other TERT-rearranged tumors were notable for higher TMBs, frequent TP53 mutations (6/7) and presence of other driver oncogenic mutations, including a concurrent fusion (TRIM24-MET). In conclusion, TERT gene rearrangements are seen in a small subset (2/55, 3.6%) of chordomas. In contrast to other TERT-rearranged tumors, where the TERT rearrangements are likely passenger events, the possibility that TERT protein overexpression representing a key event in chordoma tumorigenesis is left open.

Performance of automatic machine learning versus radiologists in the evaluation of endometrium on computed tomography.

PURPOSE: In this study, we developed radiomic models that utilize a combination of imaging features and clinical variables to distinguish endometrial cancer (EC) from normal endometrium on routine computed tomography (CT). METHODS: A total of 926 patients consisting of 416 endometrial cancer (EC) and 510 normal endometrium were included. The CT images of these patients were segmented manually, and divided into training, validation, testing and external testing sets. Non-texture and texture features of these images with endometrium or uterus as region of interest were extracted. The clinical feature "age" was also included in the feature set. Feature selection and machine learning classifier were applied to normalized feature set. This manual optimized combination was then compared with the best pipeline exported by Tree-Based Pipeline Optimization Tool (TPOT) on testing and external testing set. The performances of these machine learning pipelines were compared to that of radiologists. RESULTS: The manual expert optimized pipeline using the "reliefF" feature selection method and "Bagging" classifier on the external testing set achieved a test ROC AUC of 0.73, accuracy of 0.73 (95% CI 0.62-0.82), sensitivity of 0.64 (95% CI 0.45-0.79), and specificity of 0.78 (95% CI 0.65-0.87), while TPOT achieved a test ROC AUC of 0.79, accuracy of 0.80 (95% CI 0.70-0.87), sensitivity of 0.61 (95% CI 0.43-0.77), and specificity of 0.90 (95% CI 0.78-0.96). When compared to average radiologist performance, the TPOT achieved higher test accuracy (0.80 vs. 0.49, p < 0.001) and specificity (0.90 vs. 0.51, p < 0.001), with comparable sensitivity (0.61 vs. 0.46, p = 0.130). CONCLUSION: Our results demonstrate that automatic machine learning can distinguish EC from normal endometrium on routine CT imaging with higher accuracy and specificity than radiologists.

Encephalopathy at admission predicts adverse outcomes in patients with SARS-CoV-2 infection.

AIMS: To determine if neurologic symptoms at admission can predict adverse outcomes in patients with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). METHODS: Electronic medical records of 1053 consecutively hospitalized patients with laboratory-confirmed infection of SARS-CoV-2 from one large medical center in the USA were retrospectively analyzed. Univariable and multivariable Cox regression analyses were performed with the calculation of areas under the curve (AUC) and concordance index (C-index). Patients were stratified into subgroups based on the presence of encephalopathy and its severity using survival statistics. In sensitivity analyses, patients with mild/moderate and severe encephalopathy (defined as coma) were separately considered. RESULTS: Of 1053 patients (mean age 52.4 years, 48.0% men [n = 505]), 35.1% (n = 370) had neurologic manifestations at admission, including 10.3% (n = 108) with encephalopathy. Encephalopathy was an independent predictor for death (hazard ratio [HR] 2.617, 95% confidence interval [CI] 1.481-4.625) in multivariable Cox regression. The addition of encephalopathy to multivariable models comprising other predictors for adverse outcomes increased AUCs (mortality: 0.84-0.86, ventilation/ intensive care unit [ICU]: 0.76-0.78) and C-index (mortality: 0.78 to 0.81, ventilation/ICU: 0.85-0.86). In sensitivity analyses, risk stratification survival curves for mortality and ventilation/ICU based on severe encephalopathy (n = 15) versus mild/moderate encephalopathy (n = 93) versus no encephalopathy (n = 945) at admission were discriminative (p < 0.001). CONCLUSIONS: Encephalopathy at admission predicts later progression to death in SARS-CoV-2 infection, which may have important implications for risk stratification in clinical practice.

Chronic liver disease not a significant comorbid condition for COVID-19.

To explore the role of chronic liver disease (CLD) in COVID-19. A total of 1439 consecutively hospitalized patients with COVID-19 from one large medical center in the United States from March 16, 2020 to April 23, 2020 were retrospectively identified. Clinical characteristics and outcomes were compared between patients with and without CLD. Postmortem examination of liver in 8 critically ill COVID-19 patients was performed. There was no significant difference in the incidence of CLD between critical and non-critical groups (4.1% vs 2.9%, p = 0.259), or COVID-19 related liver injury between patients with and without CLD (65.7% vs 49.7%, p = 0.065). Postmortem examination of liver demonstrated mild liver injury associated central vein outflow obstruction and minimal to moderate portal lymphocytic infiltrate without evidence of CLD. Patients with CLD were not associated with a higher risk of liver injury or critical/fatal outcomes. CLD was not a significant comorbid condition for COVID-19.

Machine Learning-Based Prediction of COVID-19 Severity and Progression to Critical Illness Using CT Imaging and Clinical Data.

OBJECTIVE: To develop a machine learning (ML) pipeline based on radiomics to predict Coronavirus Disease 2019 (COVID-19) severity and the future deterioration to critical illness using CT and clinical variables. MATERIALS AND METHODS: Clinical data were collected from 981 patients from a multi-institutional international cohort with real-time polymerase chain reaction-confirmed COVID-19. Radiomics features were extracted from chest CT of the patients. The data of the cohort were randomly divided into training, validation, and test sets using a 7:1:2 ratio. A ML pipeline consisting of a model to predict severity and time-to-event model to predict progression to critical illness were trained on radiomics features and clinical variables. The receiver operating characteristic area under the curve (ROC-AUC), concordance index (C-index), and time-dependent ROC-AUC were calculated to determine model performance, which was compared with consensus CT severity scores obtained by visual interpretation by radiologists. RESULTS: Among 981 patients with confirmed COVID-19, 274 patients developed critical illness. Radiomics features and clinical variables resulted in the best performance for the prediction of disease severity with a highest test ROC-AUC of 0.76 compared with 0.70 (0.76 vs. 0.70, p = 0.023) for visual CT severity score and clinical variables. The progression prediction model achieved a test C-index of 0.868 when it was based on the combination of CT radiomics and clinical variables compared with 0.767 when based on CT radiomics features alone (p < 0.001), 0.847 when based on clinical variables alone (p = 0.110), and 0.860 when based on the combination of visual CT severity scores and clinical variables (p = 0.549). Furthermore, the model based on the combination of CT radiomics and clinical variables achieved time-dependent ROC-AUCs of 0.897, 0.933, and 0.927 for the prediction of progression risks at 3, 5 and 7 days, respectively. CONCLUSION: CT radiomics features combined with clinical variables were predictive of COVID-19 severity and progression to critical illness with fairly high accuracy.

Prognostication of patients with COVID-19 using artificial intelligence based on chest x-rays and clinical data: a retrospective study.

BACKGROUND: Chest x-ray is a relatively accessible, inexpensive, fast imaging modality that might be valuable in the prognostication of patients with COVID-19. We aimed to develop and evaluate an artificial intelligence system using chest x-rays and clinical data to predict disease severity and progression in patients with COVID-19. METHODS: We did a retrospective study in multiple hospitals in the University of Pennsylvania Health System in Philadelphia, PA, USA, and Brown University affiliated hospitals in Providence, RI, USA. Patients who presented to a hospital in the University of Pennsylvania Health System via the emergency department, with a diagnosis of COVID-19 confirmed by RT-PCR and with an available chest x-ray from their initial presentation or admission, were retrospectively identified and randomly divided into training, validation, and test sets (7:1:2). Using the chest x-rays as input to an EfficientNet deep neural network and clinical data, models were trained to predict the binary outcome of disease severity (ie, critical or non-critical). The deep-learning features extracted from the model and clinical data were used to build time-to-event models to predict the risk of disease progression. The models were externally tested on patients who presented to an independent multicentre institution, Brown University affiliated hospitals, and compared with severity scores provided by radiologists. FINDINGS: 1834 patients who presented via the University of Pennsylvania Health System between March 9 and July 20, 2020, were identified and assigned to the model training (n=1285), validation (n=183), or testing (n=366) sets. 475 patients who presented via the Brown University affiliated hospitals between March 1 and July 18, 2020, were identified for external testing of the models. When chest x-rays were added to clinical data for severity prediction, area under the receiver operating characteristic curve (ROC-AUC) increased from 0·821 (95% CI 0·796-0·828) to 0·846 (0·815-0·852; p<0·0001) on internal testing and 0·731 (0·712-0·738) to 0·792 (0·780-0 ·803; p<0·0001) on external testing. When deep-learning features were added to clinical data for progression prediction, the concordance index (C-index) increased from 0·769 (0·755-0·786) to 0·805 (0·800-0·820; p<0·0001) on internal testing and 0·707 (0·695-0·729) to 0·752 (0·739-0·764; p<0·0001) on external testing. The image and clinical data combined model had significantly better prognostic performance than combined severity scores and clinical data on internal testing (C-index 0·805 vs 0·781; p=0·0002) and external testing (C-index 0·752 vs 0·715; p<0·0001). INTERPRETATION: In patients with COVID-19, artificial intelligence based on chest x-rays had better prognostic performance than clinical data or radiologist-derived severity scores. Using artificial intelligence, chest x-rays can augment clinical data in predicting the risk of progression to critical illness in patients with COVID-19. FUNDING: Brown University, Amazon Web Services Diagnostic Development Initiative, Radiological Society of North America, National Cancer Institute and National Institute of Biomedical Imaging and Bioengineering of the National Institutes of Health.