RESUMEN
The conventional approach for identifying ground glass opacities (GGO) in medical imaging is to use a convolutional neural network (CNN), a subset of artificial intelligence, which provides promising performance in COVID-19 detection. However, CNN is still limited in capturing structured relationships of GGO as the texture and shape of the GGO can be confused with other structures in the image. In this paper, a novel framework called DeepChestNet is proposed that leverages structured relationships by jointly performing segmentation and classification on the lung, pulmonary lobe, and GGO, leading to enhanced detection of COVID-19 with findings. The performance of DeepChestNet in terms of dice similarity coefficient is 99.35%, 99.73%, and 97.89% for the lung, pulmonary lobe, and GGO segmentation, respectively. The experimental investigations on DeepChestNet-Lung, DeepChestNet-Lobe and DeepChestNet-COVID datasets, and comparison with several state-of-the-art approaches reveal the great potential of DeepChestNet for diagnosis of COVID-19 disease. © 2023 Wiley Periodicals LLC.
RESUMEN
Objectives: This study aims to focus on the radiological severity of covid-19 pneumonia in patients with rheumatic musculoskeletal diseases (RMD). Materials and Methods: A total of 342 Polymerase Chain Reaction positive patients were retrospectively reviewed. The patients were divided into two groups in terms of the presence of RMD. Chest Computed Tomography (CT) severity scores, demographic characteristics, hospitalization, intensive care unit (ICU) requirement, length of stay at the hospital were compared between RMD and non-RMD groups. Typical and atypical findings on CT images were identified with their incidence in both groups of patients. Results: Age and female gender were significantly higher in the RMD group (p=0.001, p=0.041). The average CT-severity score was higher in the RMD group, but the difference was not statistically significant (p=0.081). ICU transfer and mortality rates were higher in the RMD, whereas no difference was found in hospitalization rates and length of stay (p=0.002, p=0.036, p=0.280, p=0.168). Ground glass opacities, superimposed consolidation, and crazy paving patterns were the most common typical findings seen on both groups. Atypical CT findings for covid-19 pneumonia were found to be higher in the RMD group than in the non-RMD group. Conclusion: Chronic inflammation and the use of immunosuppressive drugs constitute a vulnerability to infections in RMD patients. In this study, mortality and ICU requirements were found to be higher in patients with RMD. Similarly, the higher rate of atypical chest CT findings in the RMD group may be of particular importance in the diagnosis and differential diagnosis of covid-19 pneumonia in this patient group. © 2021 Ankara Yildirim Beyazit University. All Rights Reserved.
RESUMEN
Background: To investigate the prevalence and clinical and laboratory characteristics of the cases with pulmonary embolism (PE) in the pace of coronavirus disease-2019 (COVID-19). Materials and methods: COVID-19 patients' records were retrospectively scanned from the hospital's automation system and recorded on patients' files. Results: Of 1452 COVID-19 patients, 17 (1.2%) were diagnosed with PE. Compared cases with PE with controls, it was seen that mean age was higher (p=0.036), male gender was prominent (p=0.016), patients presented with dyspnea symptoms further (p<0.001), while O2 saturation measured at room air on admission was lower (p=0.002). In PE patients, glucose (p=0.007), D-dimer (p<0.001), C-reactive protein (p<0.001) and ferritin levels (p=0.002) were higher than controls. In Receiver-Operator Characteristics analysis, the cut-off value of D-dimer in predicting PE was found to be 4211 ng/mL (p<0.001). COVID-19 patients were diagnosed with PE median five (min:max=0:36) days after hospitalization. Additionally, PE patients were found to have longer hospitalization time (p<0.001), the requirement for caring in the intensive care unit (p<0.001), and intubation (p=0.001), and non-invasive mechanical ventilation (p<0.001) in more patients, compared to controls. Mortality rates were similar in both groups, with three and 106 deaths in PE and control groups, respectively. Lower-extremity Doppler ultrasonography was performed in 196 patients, and thrombi were detected in the femoral vein in four patients, also presenting with PE. Conclusions: Even if there is no embolism without any obvious clinic of PE in all cases with COVID-19, such cases should be screened for PE in the presence of significant D-dimer elevation. © 2021 A. CARBONE Editore. All rights reserved.
RESUMEN
BACKGROUND AND STUDY AIMS: To investigate the clinical and laboratory characteristics of the cases with high lipase levels in the course of COVID-19. PATIENTS AND METHODS: Hospital records of all cases, where lipase levels were measured, and the reverse transcriptase-polymerase chain reaction test due to SARS-CoV-2 was found positive, were retrospectively investigated. Of 127 COVID-19 patients tested for lipase, 20 (15.7%) had serum lipase levels above the upper laboratory limit. The patient group with the "high lipase level" was created from these subjects, and the rest constituted the "control" group. RESULTS: While body mass index (BMI) levels were higher in the high lipase group, (p=0.014), the number of those with pre-existing diabetes mellitus (DM) was also found higher in the high lipase group than the controls (p=0.002). The history of DM was detected to increase the risk of developing high lipase level 4.63 times higher. Only two patients were diagnosed with acute pancreatitis (AP). While oxygen saturations on admission (p=0.019) and discharge (p=0.011) were lower in the high lipase group than the controls, amylase (p<0.001), C-reactive protein (CRP) (p=0.002) and D-dimer (p=0.004) levels were found higher. In addition, more patients required the treatment in intensive care unit in the high lipase group, compared to the controls (p=0.027). Accordingly, time of hospital stay became also prolonged (p=0.003). CONCLUSIONS: Pancreatic injuries or even AP may develop during SARS-CoV-2 infection, especially in those with pre-existing DM. Monitoring of pancreatic enzymes is important in COVID-19 patients, especially with pre-existing DM.