عنوان مقاله :
Rapid COVID-19 Screening Based on the Blood Test using Artificial Intelligence Methods
پديد آورندگان :
Mehralian ، Soheil Toosi University of Technology - Electrical Computer Eng. Faculty of K. N - Intelligent Systems Lab , Jalaeian Zaferani ، Effat Toosi University of Technology - Electrical Computer Eng. Faculty of K. N - Intelligent Systems Lab , Shashaani ، Shahrzad Toosi University of Technology - Electrical Computer Eng. Faculty of K. N - Intelligent Systems Lab , Kashefinishabouri ، Farnaz Toosi University of Technology - Electrical Computer Eng. Faculty of K. N - Intelligent Systems Lab , Teshnehlab ، Mohammad Toosi University of Technology - Electrical Computer Eng. Faculty of K. N - Intelligent Systems Lab , Sokhandan ، Hosein Ali BMI hospital , Dibaji Forooshani ، Zahra Sadat BMI hospital , Montazer ، Bina BMI hospital , Joneidi ، Zeinab Zanjan University of Medical Sciences - Department of genetics and molecular medicine , Vafapeyvand ، Maryam BMI hospital
كليدواژه :
Artificial intelligence , Blood test , Fuzzy system , Neural network , Support vector machine , COVID , 19 , Screen ,
چكيده فارسي :
Coronavirus Disease 2019 (COVID19) caused by the SARSCoV2 virus is spreading rapidly worldwide and has led to widespread deaths globally. As a result, the early diagnosis of patients with COVID19 is vital to control this dangerous viruschr( 39 )s release. There are two common diagnosing methods, chest computed tomography scan (CTscan) and Reverse Transcription Polymerase Chain Reaction (RTPCR) test. The most significant disadvantages of RTPCR molecular tests are the high cost and the long waiting time for test results. The common weaknesses of chest CTscan are the need for a radiologist to analyze, a misdiagnosis of flu disease due to its similarity, and risky for pregnancy and infants. This article presents a lowcost, highly available method for early detection of COVID19 based on Artificial Intelligence (AI) systems and blood tests. In this study, 6635 patientchr( 39 )s blood tests are used. Experiments conducted using three machine learning algorithms. The results show that the proposed method can detect COVID19 with an accuracy of %84 and an F1score of %83. The trained model is being used in a realworld product through an online website called CODAS.
