PSTCNN: Explainable COVID-19 diagnosis using PSO-guided self-tuning CNN

Since 2019, the coronavirus disease-19 (COVID-19) has been spreading rapidly worldwide, posing an unignorable threat to the global economy and human health. It is a disease caused by severe acute respiratory syndrome coronavirus 2, a single-stranded RNA virus of the genus Betacoronavirus. This virus is highly infectious and relies on its angiotensin-converting enzyme 2-receptor to enter cells. With the increase in the number of confirmed COVID-19 diagnoses, the difficulty of diagnosis due to the lack of global healthcare resources becomes increasingly apparent. Deep learning-based computer-aided diagnosis models with high generalisability can effectively alleviate this pressure. Hyperparameter tuning is essential in training such models and significantly impacts their final performance and training speed. However, traditional hyperparameter tuning methods are usually time-consuming and unstable. To solve this issue, we introduce Particle Swarm Optimisation to build a PSO-guided Self-Tuning Convolution Neural Network (PSTCNN), allowing the model to tune hyperparameters automatically. Therefore, the proposed approach can reduce human involvement. Also, the optimisation algorithm can select the combination of hyperparameters in a targeted manner, thus stably achieving a solution closer to the global optimum. Experimentally, the PSTCNN can obtain quite excellent results, with a sensitivity of 93.65% ± 1.86%, a specificity of 94.32% ± 2.07%, a precision of 94.30% ± 2.04%, an accuracy of 93.99% ± 1.78%, an F1-score of 93.97% ± 1.78%, Matthews Correlation Coefficient of 87.99% ± 3.56%, and Fowlkes-Mallows Index of 93.97% ± 1.78%. Our experiments demonstrate that compared to traditional methods, hyperparameter tuning of the model using an optimisation algorithm is faster and more effective. © 2023 Centro Regional de Invest. Cientif. y Tecn.. All rights reserved.

Multi-Features Disease Analysis Based Smart Diagnosis for COVID-19

Coronavirus 2019 (COVID -19) is the current global buzzword, putting the world at risk. The pandemic's exponential expansion of infected COVID-19 patients has challenged the medical field's resources, which are already few. Even established nations would not be in a perfect position to manage this epidemic correctly, leaving emerging countries and countries that have not yet begun to grow to address the problem. These problems can be solved by using machine learning models in a realistic way, such as by using computer-aided images during medical examinations. These models help predict the effects of the disease outbreak and help detect the effects in the coming days. In this paper, Multi-Features Decease Analysis (MFDA) is used with different ensemble classifiers to diagnose the disease's impact with the help of Computed Tomography (CT) scan images. There are various features associated with chest CT images, which help know the possibility of an individual being affected and how COVID-19 will affect the persons suffering from pneumonia. The current study attempts to increase the precision of the diagnosis model by evaluating various feature sets and choosing the best combination for better results. The model's performance is assessed using Receiver Operating Characteristic (ROC) curve, the Root Mean Square Error (RMSE), and the Confusion Matrix. It is observed from the resultant outcome that the performance of the proposed model has exhibited better efficient. © 2023 CRL Publishing. All rights reserved.

Association of Vitamin D and Secondary Infection in COVID-19

Introduction: Vitamin D plays a role in the modulation of cytokine release, inflammation, innate and adaptive immunity. It has been frequently discussed that the hyperinflammatory response that causes acute respiratory distress syndrome or other organ damage due to SARS-CoV-2 at the beginning of the pandemic can be modulated by the adequacy of vitamin D. The relationship of vitamin D with many conditions such as mortality, number of intensive care unit stays, disease severity, and organ damage has been investigated, but the information on its effect on secondary infections that occur during the course of the disease is limited. In this study, it was aimed to reveal the relationship of vitamin D with secondary infections that occur during the course of COVID-19 disease. Materials and Methods: Medical records of patients hospitalized in the COVID-19 pandemic service with the diagnosis of COVID-19 were evaluated retrospectively. Results: One hundred eighty-one patients were included in the study. The mean of 25(OH) vitamin D was found to be 18.76 ± 9.82 ng/mL. When 25-hydroxy vitamin D was compared with gender, disease severity, mortality, need for mechanical ventilation and presence of symptoms, no statistically significant difference was found (p> 0.05). The medical data of the patients during their hospitalization were analyzed and secondary infection was detected in 14.9% (n= 27). When 25-hydroxy vitamin D and the presence of secondary infection were compared, the 25(OH)D vitamin level of those with secondary infection was found to be low and this was found to be statistically significant (p= 0.016). As a result of the evaluation made by ROC analysis, 25-hydroxy vitamin D was found to have a diagnostic value in predicting positive culture results in COVID-19 patients (AUC= 0.771, 95% Confidence Interval= 0.612-0.810, p= 0.003, p< 0.05). Conclusion: While vitamin D continues to be an important topic of discussion in COVID-19 disease due to its effects on the immune system, it should not be forgotten that low vitamin D increases the risk of secondary infection developing in the course of COVID-19 and this may have an impact on prognosis.

Detection and Classification of COVID-19 Chest X-rays by the Deep Learning Technique

The Coronavirus disease (COVID-19) infection has become a pandemic, and this is the most critical problem that has occurred in Thailand and also expanded all over the world. As such, it is not astonishing to know that this virus has had a direct effect on hospitals with the delayed screening of patients because of the increasing number of daily cases and the shortage of medical personnel and restricted treatment space. Due to such restrictions, in this study, we used a clinical decision-making system with predictive algorithms. Predictive algorithms could potentially ease the strain on healthcare systems by identifying the diseases. Moreover, image classification is one interesting aspect of image processing. Convolutional neural network (CNN) is a widely used algorithm for image classification by separating the images of the COVID-19 disease, images with a lung infection, and normal images. To evaluate the predictive performance of our models, precision, F1-score, recall, receiver operating characteristic (ROC) curve (area under the ROC curve), and accuracy scores were used. It was observed that the predictive models trained on the laboratory findings could be used to predict the COVID-19 infection as well and could be helpful for medical experts to appropriately prioritize the resources. This could be employed to assist medical experts in validating their initial laboratory findings and could also be used for clinical prediction studies. © 2022 IEEE.

Could tear be an alternative specimen for SARS-CoV-2 detection?

Objectives: For a definitive diagnosis of COVID-19, respiratory tract samples are evaluated by polymerase chain reaction (PCR). In our study, PCR using a tear sample was used to diagnose COVID-19, and it was questioned whether it was a screening method. Unlike the general practice, Schirmer strips were used instead of a swab for tear sample collection in this study. In addition, the diagnostic values of serum procalcitonin (PCT), C-reactive protein (CRP), and Neutrophil (NEU) count in predicting COVID-19 disease from tears were also questioned. Method(s): A total of 94 patients who were positive for COVID-19 by PCR test were included in this study. Tear samples were obtained from patients with Schirmer strips, commonly used in eye examination, and studied with the PCR technique. CRP, PCT value, and NEU count were also compared between the positive and negative groups of the PCR. The obtained data were analyzed using the R Studio software, and the results were considered statistically significant for p<0.05. Result(s): Of these patients, 61 (64.9%) tear PCR was negative, and 33 (35.1%) tear PCR was positive. The mean age was 61.72 +/- 17.62 years. The patients were divided into two groups: tear PCR positive and negative. There was no significant age difference between these groups. As a result of ROC Analysis;When serum PCT, CRP, and NEU % values were examined in predicting COVID-19 disease from tears, it was seen that CRP (p=0.027) and especially PCT (p=0.003) values of patients with PCR-positive were significantly higher. Conclusion(s): PCR study on tears collected with Schirmer strips is a different and non-invasive method, but it was concluded that the proposed method could not be used as a screening test. In addition, significantly higher serum PCT values were found in patients with COVID-19 positivity in tears (p<0.05). Copyright © 2022 the author(s), published by De Gruyter.

Resilience characterized and quantified from physical activity data: A tutorial in R

Consistent physical activity is key for health and well-being, but it is vulnerable to stressors. The process of recovering from such stressors and bouncing back to the previous state of physical activity can be referred to as resilience. Quantifying resilience is fundamental to assess and manage the impact of stressors on consistent physical activity. In this tutorial, we present a method to quantify the resilience process from physical activity data. We leverage the prior operationalization of resilience, as used in various psychological domains, as area under the curve and expand it to suit the characteristics of physical activity time series. As use case to illustrate the methodology, we quantified resilience in step count time series (length = 366 observations) for eight participants following the first COVID-19 lockdown as a stressor. Steps were assessed daily using wrist-worn devices. The methodology is implemented in R and all coding details are included. For each person's time series, we fitted multiple growth models and identified the best one using the Root Mean Squared Error (RMSE). Then, we used the predicted values from the selected model to identify the point in time when the participant recovered from the stressor and quantified the resulting area under the curve as a measure of resilience for step count. Further resilience features were extracted to capture the different aspects of the process. By developing a methodological guide with a step-by-step implementation, we aimed at fostering increased awareness about the concept of resilience for physical activity and facilitate the implementation of related research. • R tutorial to quantify resilience from physical activity time series. • Physical activity resilience is measured using an idiographic approach. • Physical activity resilience is operationalized as the AUC. • Growth models are fitted to step count time series to define the limits of the AUC. • Further indicators of resilience are provided to describe the phenomenon. [ FROM AUTHOR]

The relationship between plasminogen activator inhibitor-1 levels and the course of disease in COVID-19 patients

Objectives: Studies have shown that fibrinolysis activity is insufficient in COVID-19 patients. Plasminogen activator inhibitor-1 (PAI-1) is an important antifibrinolytic molecule that plays a key role in the fibrinolytic system. In our study;we aimed to evaluate serum PAI-1 and other biochemical parameters of COVID-19 patients in terms of disease course and mortality. Method(s): A total of 40 COVID-19 patients were hospitalized in the service and intensive care unit (ICU) of our hospital from October to December 2020 and 20 healthy volunteers were included in our study. The patients were grouped as those who transferred to the ICU from the service and transferred to service from the ICU. The first and second values of the same patients in both the service and the ICU were analyzed by SPSS. Result(s): The PAI-1 levels of the patients in the ICU were significantly higher than the levels of the same patients in the service and the healthy control group (p<0.001). IL-6, ferritin, and D-dimer levels in the ICU of the same patients were significantly higher than the levels of service and healthy control group (p<0.001). A positive correlation was found between initial serum PAI-1 and D-dimer levels in patients hospitalized in the service (p=0.039) and initial serum ferritin and IL-6 levels in the ICU (p=0.031). Conclusion(s): In our study, we found that PAI-1 levels increased significantly with the increase in mortality in COVID-19 patients. Copyright © 2022 the author(s), published by De Gruyter.

The importance of LDH/Albumin, LDH/Lymphocyte, and LDH/Platelet ratios in the evaluation of COVID-19 B.1.1.7 variant

Objectives: The aim is to investigate the usefulness of lactate dehydrogenase (LDH)/Albumin, LDH/Lymphocyte and LDH/Platelet ratios on the prognosis of coronavirus disease (COVID-19) Alpha (B.1.1.7) variant pneumonia. Method(s): A total of 113 patients who were diagnosed with COVID-19 pneumonia and 60 healthy control group were included in this study. The cases were divided into 2 as classic COVID-19 group, and COVID-19 B.1.1.7 variant group. Complete blood count (CBC) and biochemical parameters of the patients were analyzed retrospectively. Patients with COVID-19 B.1.1.7 variant group were also grouped according to the length of stay in the hospital and the days of hospitalization. Result(s): LDH/Albumin, LDH/Platelet, and LDH/Lymphocyte ratios were found to be higher in COVID-19 B.1.1.7 variant group when compared to the control group (p<0.001). The ferritin, neutrophils/lymphocyte (NLR) ratio, procalcitonin (PCT) and LDH/Albumin had the highest area under the curve (AUC) values in the COVID-19 B.1.1.7 variant group (0.950, 0.802, 0.759, and 0.742, respectively). Albumin, Lymphocytes and hemoglobin values were significantly higher in the COVID-19 B.1.1.7 variant group than in the classic COVID-19 group (p<0.05). Conclusion(s): LDH/Albumin and LDH/Lymphocyte ratios may be useful for clinicians in predicting the risk of progression to pneumonia in COVID-19 B.1.1.7 variant patients. Copyright © 2022 the author(s), published by De Gruyter.

Predictive Value of Specific Cytokines for Lethal Covid-19 Outcome

In our study, we aimed to evaluate the significance of specific cytokines in blood plasma as predictive markers of COVID-associated mortality. Materials and methods. In plasma samples of 29 patients with PCR-confirmed COVID-19 we measured the concentrations of 47 molecules. These molecules included: interleukins and selected pro-inflammatory cytokines (IL-1alpha, IL-1beta, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-9, IL-12 (p40), IL-12 (p70), IL-13, IL-15, IL-17A/CTLA8, IL-17-E/IL-25, IL-17F, IL-18, IL-22, IL-27, IFNalpha2, IFNgamma, TNFalpha, TNFbeta/Lymphotoxin-alpha(LTA));chemokines (CCL2/MCP-1, CCL3/MIP-1alpha, CCL4/MIP-1beta, CCL7/MCP-3, CCL11/Eotaxin, CCL22/MDC, CXCL1/GROalpha, CXCL8/IL-8, CXCL9/MIG, CXCL10/IP-10, CX3CL1/Fractalkine);anti-inflammatory cytokines (IL-1Ra, IL-10);growth factors (EGF, FGF-2/FGF-basic, Flt-3 Ligand, G-CSF, M-CSF, GM-CSF, PDGF-AA, PDGFAB/BB, TGFalpha, VEGF-A);and sCD40L. We used multiplex analysis based on xMAP technology (Luminex, USA) using Luminex MagPix. As controls, we used plasma samples of 20 healthy individuals. Based on the results, we applied Receiver Operating Characteristic (ROC) analysis and Area Under Curve (AUC) values to compare two different predictive tests and to choose the optimal division point for disease outcome (survivors/non-survivors). To find optimal biomarker combinations, we as used cytokines concentrations as dependent variables to grow a regression tree using JMP 16 Software.Results. Out of 47 studied cytokines/chemokines/growth factors, we picked four pro-inflammatory cytokines as having high significance in evaluation of COVID-19 outcome: IL-6, IL-8, IL-15, and IL-18. Based on the results received, we assume that the highest significance in terms of predicting the outcome of acute COVID-19 belongs to IL-6 and IL-18. Conclusion. Analyzing concentrations of IL-6 and IL-18 before administering treatment may prove valuable in terms of outcome prognosis. Copyright © Arsentieva N.A. et al., 2022.

COVID-19 Diagnosis on Chest Radiograph Using Artificial Intelligence.

BACKGROUND: The coronavirus disease 2019 (COVID-19) pandemic has disrupted the world since 2019, causing significant morbidity and mortality in developed and developing countries alike. Although substantial resources have been diverted to developing diagnostic, preventative, and treatment measures, disparities in the availability and efficacy of these tools vary across countries. We seek to assess the ability of commercial artificial intelligence (AI) technology to diagnose COVID-19 by analyzing chest radiographs. MATERIALS AND METHODS: Chest radiographs taken from symptomatic patients within two days of polymerase chain reaction (PCR) tests were assessed for COVID-19 infection by board-certified radiologists and commercially available AI software. Sixty patients with negative and 60 with positive COVID reverse transcription-polymerase chain reaction (RT-PCR) tests were chosen. Results were compared against results of the PCR test for accuracy and statistically analyzed by receiver operating characteristic (ROC) curves along with area under the curve (AUC) values. RESULTS: A total of 120 chest radiographs (60 positive and 60 negative RT-PCR tests) radiographs were analyzed. The AI software performed significantly better than chance (p = 0.001) and did not differ significantly from the radiologist ROC curve (p = 0.78). CONCLUSION: Commercially available AI software was not inferior compared with trained radiologists in accurately identifying COVID-19 cases by analyzing radiographs. While RT-PCR testing remains the standard, current advances in AI help correctly analyze chest radiographs to diagnose COVID-19 infection.

Mini international neuropsychiatry interview kid or child posttraumatic stress disorder symptom scale 5I for posttraumatic stress disorder among adolescent girls infected with COVID-19.

Background: Posttraumatic stress disorder (PTSD) is a mental disorder that may develop after exposure to exceptionally life threatening or horrifying events. People suffering from PTSD are vulnerable for both physical and mental health. Objectives: To find out sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and to plot receiver operating characteristic curve taking Mini International Neuropsychiatry Interview-Kid (MINIKID) as the gold standard and Child PTSD Symptom Scale 5I (CPSS-5I) as the newer diagnostic tool for diagnosing PTSD. Materials and Methods: The cross-sectional study was carried out for a period of 6 months from January 2021 to June 2021 at R. L. Jalappa Hospital and Research Center, Kolar, Karnataka through telephonic interviews. All the data entered in Microsoft office Excel sheet, analyzed using the SPSSv22 (IBM Corp). Results: Sensitivity of the CPSS-5I was 56% and specificity was 96% compared with MINIKID. 83% and 85%, respectively, was PPV and NPV of the CPSS-5I compared with MINIKID. Area under the curve is 83.9% with P < 0.001 (72.5-95.2) indicating CPSS-5I is 84% sensitive proving to be a very good diagnostic tool for diagnosing PTSD. Furthermore, scores of 9.5 or 10.5 from CPSS-5I can be used as cutoff in diagnosing PTSD using CPSS 51. Conclusion: CPSS-5I is extremely well designed, helpful and functional tool used in diagnosing PTSD. With the current study showing CPSS-5I can be used in post-COVID PTSD diagnosis, it also provides cutoff which can be helpful in mass screening.