Evaluation of EfficientNet models for COVID-19 detection using lung parenchyma.

When the COVID-19 pandemic broke out in the beginning of 2020, it became crucial to enhance early diagnosis with efficient means to reduce dangers and future spread of the viruses as soon as possible. Finding effective treatments and lowering mortality rates is now more important than ever. Scanning with a computer tomography (CT) scanner is a helpful method for detecting COVID-19 in this regard. The present paper, as such, is an attempt to contribute to this process by generating an open-source, CT-based image dataset. This dataset contains the CT scans of lung parenchyma regions of 180 COVID-19-positive and 86 COVID-19-negative patients taken at the Bursa Yuksek Ihtisas Training and Research Hospital. The experimental studies show that the modified EfficientNet-ap-nish method uses this dataset effectively for diagnostic purposes. Firstly, a smart segmentation mechanism based on the k-means algorithm is applied to this dataset as a preprocessing stage. Then, performance pretrained models are analyzed using different CNN architectures and with our Nish activation function. The statistical rates are obtained by the various EfficientNet models and the highest detection score is obtained with the EfficientNet-B4-ap-nish version, which provides a 97.93% accuracy rate and a 97.33% F1-score. The implications of the proposed method are immense both for present-day applications and future developments.

The Impact of Early Phase COVID-19 Vaccination on Hospitalized COVID-19 Patients.

The number of cases of coronavirus disease-2019 (COVID-19) globally is over 225 million, and disease-related deaths are over 4 million. The type, prevalence, and antibody susceptibility of the virus variants of severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) and the vaccination rate and coverage are considered critical factors in the progress of COVID-19. We aimed to compare the clinical and laboratory parameters of the patients hospitalized due to COVID-19 in pre-vaccination and post-vaccination periods. We conducted this retrospective cross-sectional study in a tertiary clinic in Turkey. The files of the patients over the age of 18, whose real-time polymerase chain reaction (RT-PCR) tests were positive and who were hospitalized before (November-December 2020, Group 1) and after (March-April 2021, Group 2) COVID-19 vaccination were scanned. Patients' demographical data, clinical severity, laboratory parameters, thorax computed tomography involvement, and mortalities were recorded. The obtained data were compared among the groups. 601 patients (344 male, 57% and 257 female, 43%) were included in the study. It was observed that the patients in the Group 2 were younger (60.71 ± 14.06 vs. 66.95 ± 14.57, p < 0.001), and a significant decrease in mortality [83 (28.6%) vs.139 (44.6%), p = 0.001] were observed in Group 2. The number of patients who needed ventilatory support and the rate of pulmonary involvement was lesser in Group 2, but the difference was non-significant. C-reactive protein, D-dimer, procalcitonin levels were significantly lower in Group 2 patients. Our study shows that the age and mortality of hospitalized COVID-19 patients decreased significantly after vaccination. An increase in the number of booster doses in individuals with advanced age (age > 75) and comorbidity (especially malignancy) may contribute to the control of the disease and immunity in this population.

Do initial hematologic indices predict the severity of COVID-19 patients?

Background/aim: In this study, we aimed to evaluate the initial hematological findings analyzed on admission in confirmed COVID-19 patients who were transferred to the intensive care unit (ICU), to predict possible hematological indices. Materials and methods: Initial neutrophil to lymphocyte ratio (NLR), platelet to lymphocyte ratio (PLR), monocyte to lymphocyte ratio (MLR), red cell distribution width to platelet ratio (RPR), mean platelet volume to platelet ratio, and lymphocyte multiplied by platelet count (LYM × PLT), of 695 patients with laboratory-confirmed COVID-19 were investigated and comparisons were made between the mild/moderate and severe groups. Results: The proportion of COVID-19 cases admitted to the ICU was 3.9%. The median age of patients admitted to the ICU was significantly higher than those who were not; [68.5 (interquartile range (IQR); 21.5] years vs. 41.0 (IQR; 15.7) years; P < 0.001. Severe cases had higher NLR (6.6 vs. 2.4; P < 0.001), and MLR (0.40 vs. 0.28; P = 0.004) and lower PLR (180.0 vs. 129.0; P < 0.001) compared to that of mild or moderate patients. Among all of the parameters, the ROC curve of NLR gave us the best ability to distinguish serious patients at an early stage (AUC = 0. 819, 95% confidence interval 0.729­0.910; P < 0.001). Conclusion: These data showed that age, initial NLR, PLR, and LYM × PLT were associated with the severity of COVID-19 disease and patients' need for the ICU. Therefore, initial hemogram parameters may be essential to predict the prognosis of COVID-19 patients.