Coronavirus disease 2019 (COVID-19): survival analysis using deep learning and Cox regression model

Development and validation of predictive models for mortality of cases with COVID-19 (Omicron BA.5.2.48 and B.7.14): a retrospective study

OBJECTIVES

With the emergence of new COVID-19 variants (Omicron BA.5.2.48 and B.7.14), predicting the mortality of infected patients has become increasingly challenging due to the continuous mutation of the virus. Existing models have shown poor performance and limited clinical utility. This study aims to identify the independent risk factors and develop practical predictive models for mortality among patients infected with new COVID-19 variants.

DESIGN

A retrospective study.

SETTING AND PARTICIPANTS

We extracted data from 1029 COVID-19 patients in the respiratory disease wards of a general hospital in China between 22 December 2022 and 15 February 2023.

OUTCOME MEASURES

Mortality within 15 days after hospital discharge.

RESULTS

A total of 987 cases with new COVID-19 variants (Omicron BA.5.2.48 and B.7.14) were eventually included, among them, 153 (15.5%) died. Non-invasive ventilation, intubation, myoglobin, international normalised ratio, age, number of diagnoses, respiratory rate, pulse, neutrophil count and albumin were the most important predictors of mortality among new COVID-19 variants. The area under the curve of logistic regression (LR), decision tree (DT) and Extreme Gradient Boosting (XGBoost) models were 0.959, 0.883 and 0.993, respectively. The diagnostic accuracy was 0.926 for LR, 0.918 for DT and 0.977 for XGBoost. XGBoost model had the highest sensitivity (0.908) and specificity (0.989).

CONCLUSION

Our study developed and validated three practical models for predicting mortality in patients with new COVID-19 variants. All models performed well, and XGBoost was the best-performing model.

CTIVA: Censored time interval variable analysis

Traditionally, datasets with multiple censored time-to-events have not been utilized in multivariate analysis because of their high level of complexity. In this paper, we propose the Censored Time Interval Analysis (CTIVA) method to address this issue. It estimates the joint probability distribution of actual event times in the censored dataset by implementing a statistical probability density estimation technique on the dataset. Based on the acquired event time, CTIVA investigates variables correlated with the interval time of events via statistical tests. The proposed method handles both categorical and continuous variables simultaneously-thus, it is suitable for application on real-world censored time-to-event datasets, which include both categorical and continuous variables. CTIVA outperforms traditional censored time-to-event data handling methods by 5% on simulation data. The average area under the curve (AUC) of the proposed method on the simulation dataset exceeds 0.9 under various conditions. Further, CTIVA yields novel results on National Sample Cohort Demo (NSCD) and proteasome inhibitor bortezomib dataset, a real-world censored time-to-event dataset of medical history of beneficiaries provided by the National Health Insurance Sharing Service (NHISS) and National Center for Biotechnology Information (NCBI). We believe that the development of CTIVA is a milestone in the investigation of variables correlated with interval time of events in presence of censoring.

Survival Analysis of Hospital Length of Stay of COVID-19 Patients in Ilam Province, Iran: A Retrospective Cross-Sectional Study

Knowledge of the length of hospitalization of patients infected with coronavirus disease 2019 (COVID-19), its characteristics, and its related factors creates a better understanding of the impact of medical interventions and hospital capacities. Iran is one of the countries with the most deaths in the world (146,321 total deaths; 5 September 2023) and ranks first among the countries of the Middle East and the EMRO. Analysis of confirmed COVID-19 patients' hospital length of stay in Ilam Province can be informative for decision making in other provinces of Iran. This study was conducted to analyze the survival of COVID-19 patients and the factors associated with COVID-19 deaths in the hospitals of Ilam Province. This is a retrospective cross-sectional study. Data from confirmed COVID-19 cases in Ilam Province were obtained from the SIB system in 2019. The sample size was 774 COVID-19-positive patients from Ilam Province. Measuring survival and risk probabilities in one-week intervals was performed using Cox regression. Most patients were male (55.4%) and 55.3% were over 45 years old. Of the 774 patients, 87 (11.2%) died during the study period. The mean hospital length of stay was 5.14 days. The median survival time with a 95% confidence interval was four days. The probability of survival of patients was 80%, 70%, and 38% for 10, 20, and 30 days of hospital stay, respectively. There was a significant relationship between the survival time of patients with age, history of chronic lung diseases, history of diabetes, history of heart diseases, and hospitalization in ICU (p < 0.05). The risk of dying due to COVID-19 disease was higher among men, older age groups, and patients with a history of chronic lung diseases, diabetes, and heart disease. According to the results, taking preventive measures for elderly patients and those with underlying conditions to prevent the infection of COVID-19 patients is of potential interest. Efficiency in the management of hospital beds should also be considered.

A New Truncated Lindley-Generated Family of Distributions: Properties, Regression Analysis, and Applications

We present the truncated Lindley-G (TLG) model, a novel class of probability distributions with an additional shape parameter, by composing a unit distribution called the truncated Lindley distribution with a parent distribution function G(x). The proposed model's characteristics including critical points, moments, generating function, quantile function, mean deviations, and entropy are discussed. Also, we introduce a regression model based on the truncated Lindley-Weibull distribution considering two systematic components. The model parameters are estimated using the maximum likelihood method. In order to investigate the behavior of the estimators, some simulations are run for various parameter settings, censoring percentages, and sample sizes. Four real datasets are used to demonstrate the new model's potential.

Investigating the performance of machine learning algorithms in predicting the survival of COVID-19 patients: A cross section study of Iran

Background and Aims

Like early diagnosis, predicting the survival of patients with Coronavirus Disease 2019 (COVID-19) is of great importance. Survival prediction models help doctors be more cautious to treat the patients who are at high risk of dying because of medical conditions. This study aims to predict the survival of hospitalized patients with COVID-19 by comparing the accuracy of machine learning (ML) models.

Methods

It is a cross-sectional study which was performed in 2022 in Fasa city in Iran country. The research data set was extracted from the period February 18, 2020 to February 10, 2021, and contains 2442 hospitalized patients' records with 84 features. A comparison was made between the efficiency of five ML algorithms to predict survival, includes Naive Bayes (NB), K-nearest neighbors (KNN), random forest (RF), decision tree (DT), and multilayer perceptron (MLP). Modeling steps were done with Python language in the Anaconda Navigator 3 environment.

Results

Our findings show that NB algorithm had better performance than others with accuracy, precision, recall, F-score, and area under receiver operating characteristic curve of 97%, 96%, 96%, 96%, and 97%, respectively. Based on the analysis of factors affecting survival, heart disease, pulmonary diseases and blood related disease were the most important disease related to death.

Conclusion

The development of software systems based on NB will be effective to predict the survival of COVID-19 patients.

Deep Survival Analysis With Clinical Variables for COVID-19

OBJECTIVE

Millions of people have been affected by coronavirus disease 2019 (COVID-19), which has caused millions of deaths around the world. Artificial intelligence (AI) plays an increasing role in all areas of patient care, including prognostics. This paper proposes a novel predictive model based on one dimensional convolutional neural networks (1D CNN) to use clinical variables in predicting the survival outcome of COVID-19 patients.

METHODS AND PROCEDURES

We have considered two scenarios for survival analysis, 1) uni-variate analysis using the Log-rank test and Kaplan-Meier estimator and 2) combining all clinical variables ([Formula: see text]=44) for predicting the short-term from long-term survival. We considered the random forest (RF) model as a baseline model, comparing to our proposed 1D CNN in predicting survival groups.

RESULTS

Our experiments using the univariate analysis show that nine clinical variables are significantly associated with the survival outcome with corrected p < 0.05. Our approach of 1D CNN shows a significant improvement in performance metrics compared to the RF and the state-of-the-art techniques (i.e., 1D CNN) in predicting the survival group of patients with COVID-19.

CONCLUSION

Our model has been tested using clinical variables, where the performance is found promising. The 1D CNN model could be a useful tool for detecting the risk of mortality and developing treatment plans in a timely manner.

CLINICAL IMPACT

The findings indicate that using both Heparin and Exnox for treatment is typically the most useful factor in predicting a patient's chances of survival from COVID-19. Moreover, our predictive model shows that the combination of AI and clinical data can be applied to point-of-care services through fast-learning healthcare systems.

Physical Activity Increases Immunity to COVID-19 Infection

Coronavirus are truly one of the maximum critical fantastic-stranded non-segmented RNA viruses, named after the approximately 126-nm-diameter envelope around the nucleic acid-protein complicated. The virus causes significant harm to human fitness, including direct injury to the respiratory system, immune system compromise, worsening of the underlying clinical conditions, and eventually systemic failure and death. Exercise affects the immune system's antiviral mechanisms. Modest exercise, done before or after infection, improves morbidity and mortality to the contamination, according to animal investigations using influenza and simplex virus in the respiratory tract. Moreover, preclinical research has demonstrated that overtraining has a negative impact on the body's response to viral infections. Follow-up research has shed some light on the mechanisms underlying these discoveries. Through the activation of muscle protein synthesis, physical activity (PA) and exercise are essential for maintaining muscle mass. On the other hand, a lack of muscle contractile activity throughout the country of no exercise, particularly in elderly people, is a major contributor to anabolic rigidity and muscle atrophy.

Clinical Characteristics and Outcomes of COVID-19 Patients in Mazandaran Province, Iran

Background

The problem issue of coronaviruses is one of the most serious problems in the world. The present study aimed to investigate and describe the clinical characteristics, risk factors of fatality rate, and length of hospital stay in patients with COVID-19 in Mazandaran province.

Materials and Methods

In this epidemiological study, data from COVID-19 patients admitted to hospitals in Mazandaran province from July 22 to August 21, 2020, were reported. Multivariate logistic regression methods and the Cox proportional hazards model were used to determine the risk factors of fatality.

Results

Out of the 6759 hospitalized patients, 3111(46.03%) patients had comorbidity; 19.77% of them had diabetes, 19.97% had hypertension, and 15.28% had heart failure. Cox regression model on COVID-19 patient data showed that risk factors for fatality including having age over 60 years (HR: 1.93; P< 0.001), intubation (HR: 4.22; P<0.001), SpO2≤ 93% (HR: 2.57; P=0.006), comorbidities of cancer (HR: 1.87; P=0.006), chronic blood diseases (HR: 1.83; P=0.049), heart failure (HR: 1.63; P<0.001), and chronic kidney disease (HR: 1.98; P<0.001).

Conclusion

Paying much attention to risk factors for fatality can help identify patients with a poor prognosis in the early stages. More assessments should also be performed to examine the underlying mechanisms of these risk factors. Highlighting death-relate d risk factors is crucial to increase preparedness through appropriate medical care and prevention regulations.

Comparison of machine learning methods with logistic regression analysis in creating predictive models for risk of critical in-hospital events in COVID-19 patients on hospital admission

BACKGROUND

Machine learning (ML) algorithms have been trained to early predict critical in-hospital events from COVID-19 using patient data at admission, but little is known on how their performance compares with each other and/or with statistical logistic regression (LR). This prospective multicentre cohort study compares the performance of a LR and five ML models on the contribution of influencing predictors and predictor-to-event relationships on prediction model´s performance.

METHODS

We used 25 baseline variables of 490 COVID-19 patients admitted to 8 hospitals in Germany (March-November 2020) to develop and validate (75/25 random-split) 3 linear (L1 and L2 penalty, elastic net [EN]) and 2 non-linear (support vector machine [SVM] with radial kernel, random forest [RF]) ML approaches for predicting critical events defined by intensive care unit transfer, invasive ventilation and/or death (composite end-point: 181 patients). Models were compared for performance (area-under-the-receiver-operating characteristic-curve [AUC], Brier score) and predictor importance (performance-loss metrics, partial-dependence profiles).

RESULTS

Models performed close with a small benefit for LR (utilizing restricted cubic splines for non-linearity) and RF (AUC means: 0.763-0.731 [RF-L1]); Brier scores: 0.184-0.197 [LR-L1]). Top ranked predictor variables (consistently highest importance: C-reactive protein) were largely identical across models, except creatinine, which exhibited marginal (L1, L2, EN, SVM) or high/non-linear effects (LR, RF) on events.

CONCLUSIONS

Although the LR and ML models analysed showed no strong differences in performance and the most influencing predictors for COVID-19-related event prediction, our results indicate a predictive benefit from taking account for non-linear predictor-to-event relationships and effects. Future efforts should focus on leveraging data-driven ML technologies from static towards dynamic modelling solutions that continuously learn and adapt to changes in data environments during the evolving pandemic.

TRIAL REGISTRATION NUMBER

NCT04659187.

An Intelligent Sensor Based Decision Support System for Diagnosing Pulmonary Ailment through Standardized Chest X-ray Scans

Academics and the health community are paying much attention to developing smart remote patient monitoring, sensors, and healthcare technology. For the analysis of medical scans, various studies integrate sophisticated deep learning strategies. A smart monitoring system is needed as a proactive diagnostic solution that may be employed in an epidemiological scenario such as COVID-19. Consequently, this work offers an intelligent medicare system that is an IoT-empowered, deep learning-based decision support system (DSS) for the automated detection and categorization of infectious diseases (COVID-19 and pneumothorax). The proposed DSS system was evaluated using three independent standard-based chest X-ray scans. The suggested DSS predictor has been used to identify and classify areas on whole X-ray scans with abnormalities thought to be attributable to COVID-19, reaching an identification and classification accuracy rate of 89.58% for normal images and 89.13% for COVID-19 and pneumothorax. With the suggested DSS system, a judgment depending on individual chest X-ray scans may be made in approximately 0.01 s. As a result, the DSS system described in this study can forecast at a pace of 95 frames per second (FPS) for both models, which is near to real-time.

Development of neural network models for prediction of the outcome of COVID-19 hospitalized patients based on initial laboratory findings, demographics, and comorbidities

Background

During the process of the treatment of COVID-19 hospitalized patients, physicians still face a lot of unknowns and problems. Despite the application of the treatment protocol, it is still unknown why the medical status of a certain number of patients worsens and ends with death. Many factors were analyzed for the prediction of the clinical outcome of the patients using different methods. The aim of this paper was to develop a prediction model based on initial laboratory blood test results, accompanying comorbidities, and demographics to help physicians to better understand the medical state of patients with respect to possible clinical outcomes using neural networks, hypothesis testing, and confidence intervals.

Methods

The research had retrospective-prospective, descriptive, and analytical character. As inputs for this research, 12 components of laboratory blood test results, six accompanying comorbidities, and demographics (age and gender) data were collected from hospital information system in Sarajevo for each patient from a sample of 634 hospitalized patients. Clinical outcome of the hospitalized patients, survival or death, was recorded 30 days after admission to the hospital. The prediction model was designed using a neural network. In addition, formal hypothesis tests were performed to investigate whether there were significant differences in laboratory blood test results and age between patients who died and those who survived, including the construction of 95% confidence intervals.

Results

In this paper, 11 neural networks were developed with different threshold values to determine the optimal neural network with the highest prediction performance. The performances of the neural networks were evaluated by accuracy, precision, sensitivity, and specificity. Optimal neural network model evaluation metrics are: accuracy = 87.78%, precision = 96.37%, sensitivity = 90.07%, and specificity = 62.16%. Significantly higher values (P < 0.05) of blood laboratory result components and age were detected in patients who died.

Conclusion

Optimal neural network model, results of hypothesis tests, and confidence intervals could help to predict, analyze, and better understand the medical state of COVID-19 hospitalized patients and thus reduce the mortality rate.

Image Classification of Pests with Residual Neural Network Based on Transfer Learning

Agriculture is regarded as one of the key food sources for humans throughout history. In some countries, more than 90% of the population lives on agriculture. However, pests are regarded as one of the major causes of crop loss worldwide. Accurate and automated technology to classify pests can help pest detection with great significance for early preventive measures. This paper proposes the solution of a residual convolutional neural network for pest identification based on transfer learning. The IP102 agricultural pest image dataset was adopted as the experimental dataset to achieve data augmentation through random cropping, color transformation, CutMix and other operations. The processing technology can bring strong robustness to the affecting factors such as shooting angles, light and color changes. The experiment in this study compared the ResNeXt-50 (32 × 4d) model in terms of classification accuracy with different combinations of learning rate, transfer learning and data augmentation. In addition, the experiment compared the effects of data enhancement on the classification performance of different samples. The results show that the model classification effect based on transfer learning is generally superior to that based on new learning. Compared with new learning, transfer learning can greatly improve the model recognition ability and significantly reduce the training time to achieve the same classification accuracy. It is also very important to choose the appropriate data augmentation technology to improve classification accuracy. The accuracy rate of classification can reach 86.95% based on the combination of transfer learning + fine-tuning and CutMix. Compared to the original model, the accuracy of classification of some smaller samples was significantly improved. Compared with the relevant studies based on the same dataset, the method in this paper can achieve higher classification accuracy for more effective application in the field of pest classification.

An Ensemble Deep Learning Model with a Gene Attention Mechanism for Estimating the Prognosis of Low-Grade Glioma

While estimating the prognosis of low-grade glioma (LGG) is a crucial problem, it has not been extensively studied to introduce recent improvements in deep learning to address the problem. The attention mechanism is one of the significant advances; however, it is still unclear how attention mechanisms are used in gene expression data to estimate prognosis because they were designed for convolutional layers and word embeddings. This paper proposes an attention mechanism called gene attention for gene expression data. Additionally, a deep learning model for prognosis estimation of LGG is proposed using gene attention. The proposed Gene Attention Ensemble NETwork (GAENET) outperformed other conventional methods, including survival support vector machine and random survival forest. When evaluated by C-Index, the GAENET exhibited an improvement of 7.2% compared to the second-best model. In addition, taking advantage of the gene attention mechanism, HILS1 was discovered as the most significant prognostic gene in terms of deep learning training. While HILS1 is known as a pseudogene, HILS1 is a biomarker estimating the prognosis of LGG and has demonstrated a possibility of regulating the expression of other prognostic genes.

Factors associated with survival of Iranian patients with COVID-19: comparison of Cox regression and mixture cure model

BACKGROUNDS

SARS-CoV-2 is almost the most problematic virus of this century. It has caused extensive damage to various economic, social, and health aspects worldwide. Nowadays, coronavirus disease 2019 (COVID-19) is the most dangerous threat to human survival. Therefore, this study aimed to investigate factors associated with the survival of Iranian patients with SARS-CoV-2.

METHODS

This retrospective hospital-based cohort study was conducted on 870 COVID-19 patients with blood oxygen levels of less than 93%. Cox regression and mixture cure model were used and compared to analyze the patients' survival. It is worth noting that no similar study has been previously conducted using mixture cure regression to model the survival of Iranian patients with COVID-19.

RESULT

The cure rate and median survival time were respectively 81.5% and 20 days. Cox regression identified that respiratory distress, history of heart disease and hypertension, and older age were shown to increase the hazard. The Incidence and Latency parts of the mixture cure model respectively revealed that respiratory distress, history of hypertension, diabetes and cardiovascular diseases (CVDs), cough, fever, and older age reduced the cure odds; also, respiratory distress, history of hypertension, and CVDs, and older age increased the hazard.

CONCLUSION

The findings of our study revealed that priority should be given to older patients with a history of diabetes, hypertension, and CVDs in receiving intensive care and immunization. Also, the lower cure odds for patients with respiratory distress, fever, and cough favor early hospitalization before the appearance of severe symptoms.

Multi-Label Active Learning-Based Machine Learning Model for Heart Disease Prediction

The rapid growth and adaptation of medical information to identify significant health trends and help with timely preventive care have been recent hallmarks of the modern healthcare data system. Heart disease is the deadliest condition in the developed world. Cardiovascular disease and its complications, including dementia, can be averted with early detection. Further research in this area is needed to prevent strokes and heart attacks. An optimal machine learning model can help achieve this goal with a wealth of healthcare data on heart disease. Heart disease can be predicted and diagnosed using machine-learning-based systems. Active learning (AL) methods improve classification quality by incorporating user–expert feedback with sparsely labelled data. In this paper, five (MMC, Random, Adaptive, QUIRE, and AUDI) selection strategies for multi-label active learning were applied and used for reducing labelling costs by iteratively selecting the most relevant data to query their labels. The selection methods with a label ranking classifier have hyperparameters optimized by a grid search to implement predictive modelling in each scenario for the heart disease dataset. Experimental evaluation includes accuracy and F-score with/without hyperparameter optimization. Results show that the generalization of the learning model beyond the existing data for the optimized label ranking model uses the selection method versus others due to accuracy. However, the selection method was highlighted in regards to the F-score using optimized settings.

Evaluation of Factors Related to the Survival of Hospitalized Patients With COVID-19: Survival Analysis With Frailty Approach

OBJECTIVE

Considering that coronavirus disease 2019 (COVID-19) is an emerging disease and results in very different outcomes, from complete recovery to death, it is important to determine the factors affecting the survival of patients. Given the lack of knowledge about effective factors and the existence of differences in the outcome of individuals with similar values of the observed covariates, this study aimed to investigate the factors affecting the survival of patients with COVID-19 by the parametric survival model with the frailty approach.

METHODS

The data of 139 patients with COVID-19 hospitalized in Imam Reza Hospital in Tabriz were analyzed by the Gompertz survival model with gamma frailty effect. At first, variables with P < 0.1 in univariable analysis were included in the multivariable analysis, and then the stepwise method was used for variable selection.

RESULTS

Diabetes mellitus was significantly related to the survival of hospitalized patients (P = 0.021). The rest of the investigated variables were not significant. The frailty effect was significant (P = 0.019).

CONCLUSIONS

In the investigated sample of patients with COVID-19, diabetes was an important variable related to patient survival. Also, the significant frailty effect indicates the existence of unobserved heterogeneity that causes individuals with a similar value of the observed covariates to have different survival distributions.