Early prediction model for disease progression of COV1D-19 patients based on XGBoost: Establishment and evaluation

Objective To construct an XGBoost prediction model to predict disease severity of COVID-19 based on clinical characteristics dataset of COVID-19 patients.Methods A total of 347 laboratory-confirmed COVID-19 patients with complete medical information admitted from Feb 10 to April 5, 2020 were screened from the medical record system of Huoshenshan Hospital.Firstly, 21 features with significant differences were screened out as input features for the training model.Bayesian optimization was performed on the constructed XGBoost model to adjust the parameters, and the optimal combination of features was filtered based on feature importance.To further analyze the positive and negative effects of the numerical size of each feature on the prediction results, each feature importance was quantified and attributed by using SHapley Additive explanations (SHAP).Finally, the performance of the XGBoost prediction model was evaluated, and the model was compared and discussed with other machine learning methods, including support vector machine (SVM), naive Bayes ( NB ) , logical regression ( LR) , and k-nearest neighbors ( KNN ).Results In this study, 21 features with significant differences between the severe and non-severe groups were selected for training and validation.The optimal subset with 10 features in the k-nearest neighbor model obtained the highest value of area under curve ( AUG) among the 4 models in the validation set.XGBoost and support vector machine were better than other machine learning methods in terms of prediction performance (AUG;0.942 0, and 0.959 4 on the test set, respectively) , and the training speed of XGBoost was significantly faster.Conclusion A prediction model based on XGBoost is successfully built to achieve early prediction of disease severity of GOVID-19 patients. © 2022 Journal of Army Medical University. All rights reserved.

Epidemiological characteristics and transmission routes of COVID-19 asymptomatic infected patients in Shaanxi Province

Objective: To assess the epidemiological characteristics and transmission risk of asymptomatic COVID-19 infection in Shaanxi Province. Methods: A dense population database of COVID-19 cases in Shaanxi Province was constructed as of March 26, 2020, and 28 asymptomatic infected patients were selected through case epidemiological investigation report for descriptive analysis. Results: In Shaanxi Province, the majority of asymptomatic COVID-19 infections were in the 20-59 years old group, and women took up a higher proportion than men. 82.14% of asymptomatic infections were found in the centralized isolation state. Nearly 80% of asymptomatic infections and confirmed cases were mainly exposed by living together in the family. The median number of days from last contact with the source of infection for all asymptomatic infected persons was 13, with 3 (10.71%) showing clinical symptoms. Epidemiological investigations showed that asymptomatic infections could spread as a result of shared family life. Conclusion: At present, imported cases in Shaanxi Province continue to exist. Considering the concealed transmission of asymptomatic infections, prevention and control work is still facing challenges. © 2021, Editorial Board of Journal of Xi'an Jiaotong University (Medical Sciences). All right reserved.

Analysis of the path from discovery of novel coronavirus disease 2019 cases to its diagnosis in Shaanxi Province

Objective: To explore the problems and experience in the prevention and control of novel coronavirus disease 2019 (COVID-19) by analyzing the path from its discovery to its diagnosis in Shaanxi Province. Methods: We collected the important data of COVID-19 epidemic situation in Shaanxi Province as of March 26, 2020, the trajectory of the diagnosed cases, contact history, close contacts and other important information for a descriptive epidemiological analysis. Results: Totally 72.65% (178/245) of the confirmed COVID-19 cases in Shaanxi Province were not in isolation or management at the time of onset, which resulted in an infection rate of 1.37% in close-knit populations. The duration between the onset of local COVID-19 cases in Shaanxi and the delay in confirmed diagnosis was relatively long (6.30 days), and the mean incubation period was 7.27 days. The delay time (7.29 days) between onset and diagnosis of preepidemic cases was longer than that (6.19 days) of later epidemic cases. Conclusion: According to the epidemic situation, we should gradually sort out the loopholes in previous work and at the same time make timely evaluation of the new problems arising under the new situation and constantly adjust the prevention and control strategies. © 2020, Editorial Board of Journal of Xi'an Jiaotong University (Medical Sciences). All right reserved.

[Typical case analysis of COVID-19 cluster epidemic in Shaanxi, 2020].

Objective: By analyzed the transmission patterns of 4 out of the 51 COVID-19 cluster cases in Shaanxi province to provide evidences for the COVID-19 control and prevention. Methods: The epidemiological data of RT-PCR test-confirmed COVID-19 cases were collected. Transmission chain was drawn and the transmission process was analyzed. Results: Cluster case 1 contained 13 cases and was caused by a family of 5 who traveled by car to Wuhan and returned to Shaanxi. Cluster case 2 had 5cases and caused by initial patient who participated family get-together right after back from Wuhan while under incubation period. Cluster case 3 contained 10 cases and could be defined as nosocomial infection. Cluster case 4 contained 4 cases and occurred in work place. Conclusion: Higher contact frequency and smaller places were more likely to cause a small-scale COVID-19 cluster outbreak, with potential longer incubation period. COVID-19 control strategies should turn the attention to infection prevention and control in crowded places, management of enterprise resumption and prevention of nosocomial infection.

[Analysis of transmission characteristics of COVID-19 in Shaanxi Province].

A total of 245 cases of COVID-19 in Shaanxi Province reported in the China information system for disease control and prevention as of February 24, 2020 were selected as the research objects, the cases are divided into imported cases (116 cases, 47.3%) and local cases (129 cases, 52.7%), their basic characteristics, time distribution, transmission mode, intergenerational interval and latent period transmission are analyzed. The age of local cases [(51.74±15.67) years old], female patients (69 cases, 53.5%), housework and retired staff (40 cases, 31.0%), and patients isolated at the time of onset (50 cases, 38.8%) were higher than imported cases, respectively[(40.66±15.41) years old, (45 cases, 38.8%), (21 cases, 18.1%), (17 cases, 14.6%)] (P values were < 0.05); The infection rate was 0.8% (31/3 666) in close contacts with local cases, which was lower than imported cases 2.0% (69/3 435) (P<0.001); The main source of infection in local cases was relatives (70 cases, 54.3%), and the main way of infection was living together and party (90 cases, 69.8%); the proportion of latent period transmission in our province was 15.5% (20 cases), and the interval between the second-generation case and the source of infection was about 4 days, and the interval between generations was about 6 days. In summary, the main way of infection of local cases in Shaanxi Province was living together and party, there were a certain proportion of latent period transmission cases at present, it's suggested that the investigation of close contacts should be started 4 days or earlier before the onset of the case.

[Epidemiological characteristics of COVID-19 in Shaanxi province].

Objective: To understand the incidence trend and epidemiological characteristics of COVID-19 in Shaanxi province. Methods: The incidence data of COVID-19 reported in Shaanxi as of 22 February, 2020 were collected for an epidemiological descriptive analysis. Results: A total of 245 confirmed cases of COVID-19 were reported in Shaanxi. Most cases were mild (87.76%). As time passed, the areas where confirmed cases were reported continued to increase. The case number in Xi'an was highest, accounting for nearly half of the total reported cases in the province. The epidemic pattern in Shaanxi had gradually shifted from imported case pattern to local case pattern, and the transmission of local cases was mainly based on family cluster transmission. The confirmed cases from different sources had caused the secondary transmission in Shaanxi. After February 7, the number of reported cases began to fluctuate and decrease stably, indicating a decrease-to-zero period. Conclusions: At present, the overall epidemic of COVID-19 in Shaanxi has gradually been mitigated. However, considering the approaching of return to work and study and the increasing of imported cases from other countries, the prevention and control of COVIS-19 in Shaanxi will face new challenges.