ABSTRACT
Background: Given that 2019 novel coronavirus (COVID-19) spreads rapidly, it is critical to make rapid and accurate detection of COVID-19 patients towards containment of SARS-CoV-2 virus. At present, COVID-19 patients are mainly identified through viral nuclear acid testing (NAT). However, factors such as time for patients being tested, experience of test operators, and specimen’s preparation, might affect the accuracy of testing results. The purpose of this study was to use different classification and feature selection methods to improve the diagnostic accuracy of COVID-19 patients. Methods: We utilized seven machine learning algorithms for assisting diagnosis of COVID-19 by developing a non-NAT algorithm. In order to reduce the number of input features while maintaining the models’ performance so as to decrease the cost and time consumption, we adopted three algorithms, such as Chi-square test, variance analysis, and feature importance tests to identify the optimal feature sets. Findings: The XGBoost and RF models displayed the best performance for COVID-19 detection, with the highest accuracy rate more than 0·96. The accuracy of RF model was 0·968 when using only ten hematological features and body temperature. Interpretation: Ten blood features and body temperature can fairly accurately determine whether a suspected patient is infected with COVID-19. Our model can improve the diagnostic accuracy of COVID-19 and reduce the spread. Funding: This work is supported by grants from the National Key Research and Development Program of China under Grant 2017YFE0123600, the Natural Science Foundation of China (81873931, 81974382 and 81773104), the Frontier Exploration Program of Huazhong University of Science and Technology (2015TS153), and the Major Scientific and Technological Innovation Projects in Hubei Province (2018ACA136).Declaration of Interests: All the authors stated that the paper had never been published elsewhere, and that there were no competing economic interests.Ethics Approval Statement: The collection, use, and retrospective analysis of chest CT images, CFs and SARS-CoV-2 nucleic acid PCR results of patients were approved by the institutional ethical committees of HUST-UH (IRB ID: [2020] IEC(A001)).
Subject(s)
COVID-19ABSTRACT
Background: Coronavirus disease 19 (COVID-19) caused by a novel coronavirus (2019-nCoV) has now been classified as a pandemic by the World Health Organization (WHO). It has been reported that 2019-nCoV could be detected in various human secreta and excreta such as sputum, faeces and urine. However, the detection of 2019-nCoV in saliva has not been reported to date. To provide a more comprehensive understanding of the effects of 2019-nCoV on oral health and possible saliva transmission, the expression of the putative ACE2 (angiotensin-converting enzyme II) receptor of 2019-nCoV in salivary gland epithelial cells were analyzed in bulk RNA-seq profiles from public databases. Additionally, to evaluate the oral health status of COVID-19 patients, a questionnaire survey on oral-related symptoms of COVID-19 patients was performed.Methods: To analyze ACE2 expression in salivary glands, bulk RNA-seq profiles from four public datasets including GTEx dataset, HPA dataset, FANTOM5 dataset and Consensus dataset were collected. Altogether, 31 COVID-19 patients were recruited, whose 2019-nCoV nucleic acid detection remained positive before or on the day of sample collection. Of these 31 patients, there were 4 critically-ill cases whereby oropharyngeal swabs were tested positive. Saliva was collected from the opening of the salivary gland canal of cleaned oral cavity to avoid contamination by other secretions from the respiratory tract. At the same time, oropharyngeal swabs were also collected. Then the presence of 2019-nCoV nucleic acids in saliva was detected by RT-PCR. Additionally, a questionnaire survey on various oral symptoms such as dry mouth and amblygeustia was also carried out on COVID-19 patients.Findings: ACE2 expression was present at detectable levels in the salivary glands. Interestingly, we found 4 cases with positive detection of 2019-nCoV nucleic acid in saliva. It is worth noting that 3 cases with positive detection in saliva were critically-ill patients on ventilator support, thus implying a high potential (75%) for detection of 2019-nCoV in the saliva of critically-ill patients. At the same time, the two major oral-related symptoms, dry mouth (46.3%) and amblygeustia (47.2%), were manifested by a relatively high proportion of 108 COVID-19 patients.Interpretation: This study thus confirms the expression of ACE2 in the salivary glands, and demonstrates the possibility of 2019-nCoV infection of the salivary glands. The reason that the positive saliva detection rate was as high as 75% (3/4) in critically-ill patients in our study might be due to virus invasion caused by high viral loads or destroyed salivary glands at the late stage of the disease. The emergence of viral particles in the saliva might be an indication that the disease condition of the patient has deteriorated and progressed to the terminal stage. Hence, saliva may be a new source of diagnostic specimens for critically-ill patients, since it can be easily collected without any invasive procedures.The two major oral-related symptoms, dry mouth and amblygeustia, were manifested by a relatively high proportion of COVID-19 patients suggesting that oral symptoms can also be considered as initial symptoms of COVID-19 infection.