Development of an Artificial Intelligence Method to Detect COVID-19 Pneumonia in Computed Tomography Images

Introduction: This study aimed to construct an artificial intelligence system to detect Coronavirus disease-2019 (COVID-19) pneumonia on computed tomography (CT) images and to test its diagnostic performance. Methods: Data were acquired between March 18-April 17, 2020. CT data of 269 reverse tran-scriptase-polymerase chain reaction proven patients were extracted, and 173 studies (122 for training, 51 testing) were finally used. Most typical lesions of COVID-19 pneumonia were la-beled by two radiologists using a custom tool to generate multiplanar ground-truth masks. Us-ing a patch size of 128x128 pixels, 18,255 axial, 71,458 coronal, and 72,721 sagittal patches were generated to train the datasets with the U-Net network. Lesions were extracted in the or-thogonal planes and filtered by lung segmentation. Sagittal and coronal predicted masks were reconverted to the axial plane and were merged into the intersect-ed axial mask using a voting scheme. Results: Based on the axial predicted masks, the sensitivity and specificity of the model were found as 91.4% and 99.9%, respectively. The total number of positive predictions has increased by 3.9% by the use of intersected predicted masks, whereas the total number of negative predic-tions has only slightly decreased by 0.01%. These changes have resulted in 91.5% sensitivity, 99.9% specificity, and 99.9% accuracy. Conclusion: This study has shown the reliability of the U-Net architecture in diagnosing typical pulmonary lesions of COVID-19 in CT images. It also showed a slightly favorable effect of the intersection method to increase the model's performance. Based on the performance level pre-sented, the model may be used in the rapid and accurate detection and characterization of the typical COVID-19 pneumonia to assist radiologists.

Covid-19 Related Fatality and Risk Factors in Multiple Myeloma: A Multicenter Cohort Study

The distinctive clinical course and outcomes of COVID-19 infection in multiple myeloma patients are still not well established. In this study, we aimed to assess the clinical outcomes and associated factors of COVID-19 in patients with multiple myeloma (MM). This is a multi-center retrospective cohort study. Multiple myeloma patients treated in two tertiary centers were investigated, and the patients diagnosed with COVID-19 during follow-up were included. The main characteristics and clinical outcomes of patients were analyzed. A total of thirty patients were included for analysis. In this cohort, autologous hematopoietic stem cell transplantation (AHSCT) was performed in 63.3% of the patients, and 36.7% were in complete remission when COVID-19 was detected. The total fatality rate (FR) was 36%, and the COVID-19-related case fatality rate (CFR) was 30% for MM patients in our cohort. There was two non-COVIDrelated mortality. The CFR was associated with intensive care unit admission (26.7%, p< 0.001), mechanical ventilation (26.6%, p< 0.001), increased lactate dehydrogenase (p= 0.008) and lymphopenia (p= 0.042). Older age (> 65-years), stem cell transplantation, and comorbidities were not effective on the fatality rate. This study shows that the CFR rate was high in MM patients, irrespective of AHSCT status. Therefore, we suggest strict monitoring and adequate vaccination in this group. However, further studies, including vaccination data with a larger group of patients, are needed to clarify the literature.

Preference and perception of treatment choices between radiologists and urologists for treatment for urgent urinary obstruction – nephrostomy versus retrograde ureteric stent: European survey from YAU and ESUT Endourology groups

Introduction & Objectives: Urological emergencies related to urinary obstruction need Percutaneous Nephrostomy (PCN) or Retrograde Ureteric Stent (RUS). The choice of treatment is often debated between radiologists and urologists due to differences in perception for given scenarios and the skill set needed for these. We wanted to conduct a European survey to determine the preference of treatment in different clinical situations. Materials & Methods: A European survey was conducted via the EAU sections (YAU and ESUT) for preference and treatment choices between radiologists and urologists for using PCN or RUS or primary ureteroscopy (URS) in various clinical scenarios. Responders were asked to select urinary drainage for 3 clinical scenarios before and after reading evidence from literature on use of PCN or RUS. The scenarios were ureteric stone related – infected obstructed kidney (scenario 1), obese patient with pain and hydronephrosis (scenario 2) and solitary kidney with deranged renal function (scenario 3). Results: Of the responses (n=367), there were 15.4% (n=57) radiologists and 310 (84.5%) urologists. The choice of drainage for scenario 1,2 and 3 between urologists and radiologists pre- and post-evidence perusal are shown in Table 1. Regarding QoL, cost and radiation dose (Table 2), the perception was that Radiologists appear to consider JJ stents to provide a better QoL (p=0.0004) and more radiation exposure (p<0.0001) than Urologists. The perception in both groups was that stent was more expensive (p=0.652507). With COVID-19 pandemic, there was also a rise in the usage of local anaesthetic stent and URS procedures. (Table Presented) Conclusions: Choice of urinary drainage for urological emergencies is dependent on multiple factors, but prompt management is paramount. This survey has shown how urologists and radiologists both put patient safety at the forefront and their choice of treatment reflects their expertise in the given technique.