Prediction of diagnosis and prognosis of COVID-19 disease by blood gas parameters using decision trees machine learning model: a retrospective observational study.
{"title":"Prediction of diagnosis and prognosis of COVID-19 disease by blood gas parameters using decision trees machine learning model: a retrospective observational study.","authors":"Mehmet Tahir Huyut, Hilal Üstündağ","doi":"10.4103/2045-9912.326002","DOIUrl":null,"url":null,"abstract":"<p><p>The coronavirus disease 2019 (COVID-19) epidemic went down in history as a pandemic caused by corona-viruses that emerged in 2019 and spread rapidly around the world. The different symptoms of COVID-19 made it difficult to understand which variables were more influential on the diagnosis, course and mortality of the disease. Machine learning models can accurately assess hidden patterns among risk factors by analyzing large-datasets to quickly predict diagnosis, prognosis and mortality of diseases. Because of this advantage, the use of machine learning models as decision support systems in health services is increasing. The aim of this study is to determine the diagnosis and prognosis of COVID-19 disease with blood-gas data using the Chi-squared Automatic Interaction Detector (CHAID) decision-tree-model, one of the machine learning methods, which is a subfield of artificial intelligence. This study was carried out on a total of 686 patients with COVID-19 (n = 343) and non-COVID-19 (n = 343) treated at Erzincan-Mengücek-Gazi-Training and Research-Hospital between April 1, 2020 and March 1, 2021. Arterial blood gas values of all patients were obtained from the hospital registry system. While the total-accuracyratio of the decision-tree-model was 65.0% in predicting the prognosis of the disease, it was 68.2% in the diagnosis of the disease. According to the results obtained, the low ionized-calcium value (< 1.10 mM) significantly predicted the need for intensive care of COVID-19 patients. At admission, low-carboxyhemoglobin (< 1.00%), high-pH (> 7.43), low-sodium (< 135.0 mM), hematocrit (< 40.0%), and methemoglobin (< 1.30%) values are important biomarkers in the diagnosis of COVID-19 and the results were promising. The findings in the study may aid in the early-diagnosis of the disease and the intensive-care treatment of patients who are severe. The study was approved by the Ministry of Health and Erzincan University Faculty of Medicine Clinical Research Ethics Committee.</p>","PeriodicalId":18559,"journal":{"name":"Medical Gas Research","volume":"12 2","pages":"60-66"},"PeriodicalIF":3.0000,"publicationDate":"2022-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ftp.ncbi.nlm.nih.gov/pub/pmc/oa_pdf/c1/cd/MGR-12-60.PMC8562394.pdf","citationCount":"29","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Medical Gas Research","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.4103/2045-9912.326002","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MEDICINE, RESEARCH & EXPERIMENTAL","Score":null,"Total":0}
引用次数: 29
Abstract
The coronavirus disease 2019 (COVID-19) epidemic went down in history as a pandemic caused by corona-viruses that emerged in 2019 and spread rapidly around the world. The different symptoms of COVID-19 made it difficult to understand which variables were more influential on the diagnosis, course and mortality of the disease. Machine learning models can accurately assess hidden patterns among risk factors by analyzing large-datasets to quickly predict diagnosis, prognosis and mortality of diseases. Because of this advantage, the use of machine learning models as decision support systems in health services is increasing. The aim of this study is to determine the diagnosis and prognosis of COVID-19 disease with blood-gas data using the Chi-squared Automatic Interaction Detector (CHAID) decision-tree-model, one of the machine learning methods, which is a subfield of artificial intelligence. This study was carried out on a total of 686 patients with COVID-19 (n = 343) and non-COVID-19 (n = 343) treated at Erzincan-Mengücek-Gazi-Training and Research-Hospital between April 1, 2020 and March 1, 2021. Arterial blood gas values of all patients were obtained from the hospital registry system. While the total-accuracyratio of the decision-tree-model was 65.0% in predicting the prognosis of the disease, it was 68.2% in the diagnosis of the disease. According to the results obtained, the low ionized-calcium value (< 1.10 mM) significantly predicted the need for intensive care of COVID-19 patients. At admission, low-carboxyhemoglobin (< 1.00%), high-pH (> 7.43), low-sodium (< 135.0 mM), hematocrit (< 40.0%), and methemoglobin (< 1.30%) values are important biomarkers in the diagnosis of COVID-19 and the results were promising. The findings in the study may aid in the early-diagnosis of the disease and the intensive-care treatment of patients who are severe. The study was approved by the Ministry of Health and Erzincan University Faculty of Medicine Clinical Research Ethics Committee.
期刊介绍:
Medical Gas Research is an open access journal which publishes basic, translational, and clinical research focusing on the neurobiology as well as multidisciplinary aspects of medical gas research and their applications to related disorders. The journal covers all areas of medical gas research, but also has several special sections. Authors can submit directly to these sections, whose peer-review process is overseen by our distinguished Section Editors: Inert gases - Edited by Xuejun Sun and Mark Coburn, Gasotransmitters - Edited by Atsunori Nakao and John Calvert, Oxygen and diving medicine - Edited by Daniel Rossignol and Ke Jian Liu, Anesthetic gases - Edited by Richard Applegate and Zhongcong Xie, Medical gas in other fields of biology - Edited by John Zhang. Medical gas is a large family including oxygen, hydrogen, carbon monoxide, carbon dioxide, nitrogen, xenon, hydrogen sulfide, nitrous oxide, carbon disulfide, argon, helium and other noble gases. These medical gases are used in multiple fields of clinical practice and basic science research including anesthesiology, hyperbaric oxygen medicine, diving medicine, internal medicine, emergency medicine, surgery, and many basic sciences disciplines such as physiology, pharmacology, biochemistry, microbiology and neurosciences. Due to the unique nature of medical gas practice, Medical Gas Research will serve as an information platform for educational and technological advances in the field of medical gas.