COVID-19 Mortality Risk Prediction using Clinical and Laboratory Examination: Machine Learning Approach for Implementation (preprint)

Background and Aim: We aimed to propose a mortality risk prediction tool to facilitate COVID-19 patient management and allocation for the frontline physician on admission day. Methods: We used a dataset of confirmed COVID-19 patients admitted to three general hospitals in Tehran. Clinical and laboratory values on admission were gathered. Different machine learning methods were used to assess the risk of in-hospital mortality, including logistic regression, k-nearest neighbor (KNN), gradient boosting classifier, random forest, support vector machine, and deep neural network (DNN). Least absolute shrinkage and selection operator (LASSO) regression and Boruta feature selection methods were used for feature selection. The proposed model was selected using the area under the receiver operator curve (AUC). Furthermore, a dataset from the fourth hospital was used for external validation. Results: 5320 hospitalized COVID-19 patients were enrolled in the study with a mean age of 61.6± 17.6 years and a fatality rate of 17.24% (N=917). All methods showed fair performance with AUC>80%, except for the KNN method. The feature selection method selected ten laboratories and eight clinical features. Our proposed DNN and LASSO feature selection methods showed AUC scores of 83.4% and 82.8% in internal and external validation, respectively. Furthermore, our imputer worked fairly when two out of ten laboratory parameters were missing (AUC=81.8%). Conclusion: We worked intimately with healthcare professionals to provide a tool that can solve real-world needs. Our proposed model showed promising results and confirms the potential of ML methods for use in clinical practice as a decision-support system. Future studies are warranted to investigate barriers to the implementation of ML tools.

Visual Transformer and Deep CNN Prediction of High-risk COVID-19 Infected Patients using Fusion of CT Images and Clinical Data (preprint)

Despite the globally reducing hospitalization rates and the much lower risks of Covid-19 mortality, accurate diagnosis of the infection stage and prediction of outcomes are clinically of interest. Advanced current technology can facilitate automating the process and help identifying those who are at higher risks of developing severe illness. Deep-learning schemes including Visual Transformer and Convolutional Neural Networks (CNNs), in particular, are shown to be powerful tools for predicting clinical outcomes when fed with either CT scan images or clinical data of patients. This paper demonstrates how a novel 3D data fusion approach through concatenating CT scan images with patients clinical data can remarkably improve the performance of Visual Transformer and CNN models in predicting Covid-19 infection outcomes. Here, we explore and represent comprehensive research on the efficiency of Video Swin Transformers and a number of CNN models fed with fusion datasets and CT scans only vs a set of conventional classifiers fed with patients clinical data only. A relatively large clinical dataset from 380 Covid-19 diagnosed patients was used to train/test the models. Results show that the 3D Video Swin Transformers fed with the fusion datasets of 64 sectional CT scans+67 (or 30 selected) clinical labels outperformed all other approaches for predicting outcomes in Covid-19-infected patients amongst all techniques (i.e., TPR=0.95, FPR=0.40, F0.5 score=0.82, AUC=0.77, Kappa=0.6). Results indicate possibilities of predicting the severity of outcome using patients CT images and clinical data collected at the time of admission to hospital.

COVID-19 vaccines breakthrough infection and adverse reactions in medical students: A nationwide survey in Iran (preprint)

Purpose This study aimed to investigate the rate of COVID-19 breakthrough infection and adverse events in medical students.Methods Iranian medical students receiving two doses of COVID-19 vaccines were included in this retrospective cohort study. The medical team gathered the demographic characteristics, comorbidities, type of vaccine, adverse events following vaccination, and history of COVID-19 infection data through a phone interview. The frequency of adverse events and breakthrough infection was stratified by vaccine type (ChAdOx1-S, Gam-COVID-Vac, and BIBP-CorV).Results A total of 3591 medical students enrolled in this study, of which 57.02% were females, with a mean age of 23.31 + 4.87. A PCR-confirmed and suspicious-for-COVID-19 breakthrough infection rate of 4.51% and 7.02% was detected, respectively. There was no significant relation between breakthrough infection and gender, BMI, blood groups, and comorbidities. However, there was a significant difference in breakthrough infection rate among different types of vaccines (P = 0.001) and history of COVID-19 infection (P = 0.001). A total of 16 participants were hospitalized for COVID-19 infection, and no severe infection or death was observed in the studied population.Conclusion Vaccination prevented severe COVID-19 infection, although a high breakthrough infection rate was evident among Iran medical students during the Delta variant’s peak. Vaccine effectiveness may be fragile during emerging new variants and in high-exposure settings. Moreover, adverse events are rare, and the benefits of vaccination outweigh the side effects. However, many limitations challenged this study, and the results should be cautious.

The comparison of Post-Operative Complications pre Covid era versus during Covid-Era based on Clavien-Dindo-classification: A Systematic Review and Meta-Analysis (preprint)

Introduction: Coronavirus Disease 2019 (COVID -19) pandemic challenged the healthcare system drastically, and it was concomitant with a remarkable decline in surgeries and modified routine care of patients worldwide. This systematic review and meta-analysis aimed to compare the surgical complications before COVID -19 (Pre-COVID) and after COVID -19 (post-COVID) appearance using the Clavien-Dindo classification (CDC). Methods: between January 1, 2019, to November 3, 2021, we performed a comprehensive search in PubMed/Medline and Scopus for studies reporting the postoperative complications based on/transformable to CDC. Result: From 909 screened articles, 34 studies were included for systematic review. Among included articles, 11 were eligible for meta-analysis. Nineteen thousand one hundred thirty-seven patients (pre-COVID: 3522, post-COVID: 15615) were included, mostly undergoing elective surgeries (86.32%). According to CDC classification, there were no significant change between pre-COVID and post-COVID for grade 1 (Odds ratio (OR) and 95% confidence interval (95-CI): 0.99, 0.60-1.63, p=0.96), grade 2 (OR and 95-CI: 0.65, 0.42-1.01, p = 0.055), grade 3 (OR and 95-CI: 0.86, 0.48-1.57, p=0.64), grade 4 (OR and 95-CI: 0.85, 0.46-1.57, p =0.60). However, the postoperative mortality was lower before the COVID -19 outbreak (OR and 95-CI: 0.51, 0.27-0.95, p= 0.035). The included studies for systematic review and meta-analysis had a low risk of bias and unsignificant publication bias. Conclusion: Although delivering routine surgery was challenging, the postoperative complications during the pandemic remained identical to the pre-pandemic era. The stricter patient selection tending to choose more critical states and more advanced clinical stages of the operated patients may explain some extent of higher mortality during the pandemic. Adopting preventive strategies helped deliver surgeries during the outbreak of COVID -19 while limiting the capacity of operations and admissions

Excess All-Cause Mortality and COVID-19 Reported Fatality in Iran (April 2013 - September 2021): Age and Sex Disaggregated Time Series Analysis (preprint)

Objective: The actual impact of the pandemic on COVID-19 specific mortality is still unclear due to the variability in access to diagnostic tools. This study aimed to estimate the excess all-cause mortality in Iran until September 2021 based on the national death statistics.Results: The autoregressive integrated moving average used to predict seasonal all-cause death in Iran (R-squared=0.45). We observed a 38.8% (95% confidence interval (CI): 29.7%-40.1%) rise in the all-cause mortality from 22 June 2020 to 21 June 2021. The excess all-cause mortality per 100,000 population were 178.86 (95% CI: 137.2 - 220.5, M:F ratio = 1.3) with 49.1% of these excess deaths due to COVID-19. Comparison of spring 2019 and spring 2021 revealed that the highest percent increase in mortality was among men, aged 65-69 years old (77%) and women, aged 60-64 years old (86.8%). Moreover, the excess mortality among 31 provinces of Iran ranged from 109.7 (Hormozgan) to 273.2 (East-Azerbaijan) per 100,000 population. In conclusion, there was a significant rise in all-cause mortality during the pandemic. Since COVID-19 fatality explains about half of this rise, the rise in other causes of death and underestimation in reported data should be concerned by further studies.