Developing a decision tree model for screening SARS-CoV-2 infection
Keywords:
Coronavirus 2019, COVID-19, COVID-19 Antigen, Rapid antigen test, Decision treeAbstract
The development of a decision tree model using Rapid Antigen Tests (RAT) combined with clinical data aims to determine the sensitivity and specificity of SARS-CoV-2 detection through the RAT method and compare with Realtime RT-PCR method. This study also seeks to assess the sensitivity and specificity of the decision tree model. The findings indicate that the RAT for detecting the antigens of the 2019 coronavirus has a sensitivity of 91.14% (CI 88.48 - 95.45), specificity of 99.21% (CI 98.10 - 99.64), positive predictive value of 97.75% (CI 95.77 - 98.81), negative predictive value of 96.74% (CI 95.64 - 97.79), and accuracy of 97.00% (CI 96.03 - 97.79).
The study evaluates the performance of three classification algorithms using decision tree techniques: J48, ID3, and CART. The results indicate that the most suitable primary model for application is the J48 classifier combined with the SMOTE technique, yielding a sensitivity of 96.1%, specificity of 99.2%, and accuracy of 98.5% after pruning to reduce model complexity. This model produced eight classification rules. When the final model was evaluated against blind data, it achieved sensitivity, specificity, and accuracy rates of 96.2%, 98.7%, and 97.4%, respectively. Therefore, RAT is an efficient method of detecting antigens of coronavirus due to its quick results and suitability for highly affected areas. Additionally, the integration of clinical data with RAT analysis can improve the screening process and increase the sensitivity and accuracy of the test results.
References
Li Q, Guan X, Wu P, Wang X, Zhou L, Tong Y, et al. Early transmission dynamics in Wuhan, China,
of novel coronavirus–infected pneumonia. N Engl J Med. 2020;382(13):1199-207.
Emergency Operations Center, Department of Disease Control. The Coronavirus Disease 2019 Situation [Internet]. 2021 [cited 2021 May 14]. Available from: https://ddc.moph.go.th/viralpneumonia/file/situation/situation-no92-040463.pdf
Emergency Operations Center, Department of Disease Control. The Coronavirus Disease 2019 Situation [Internet]. 2021 [cited 2021 May 5]. Available from: https://ddc.moph.go.th/viralpneumonia/file/situation/situation-no372-090164.pdf
Department of Medical Sciences, Ministry of Public Health. Manual of Laboratory Diagnostic for coronavirus disease 2019 (COVID-19). Bangkok: Text and Journal Publication company Limited; 2021. p. 7-28.
Bruzzone B, De Pace V, Caligiuri P, Ricucci V, Guarona G, Pennati BM, et al. Comparative diagnostic performance of rapid antigen detection tests for COVID-19 in a hospital setting. Int J Infect Dis. 2021;107:215-8.
Hayer J, Kasapic D, Zemmrich C. Real-world clinical performance of commercial SARS-CoV-2 rapid antigen tests in suspected COVID-19: A systematic meta-analysis of available data as of November 20, 2020. Int J Infect Dis. 2021;108:592-602.
Porte L, Legarraga P, Vollrath V, Aguilera X, Munita JM, Araos R, et al. Evaluation of a novel antigen-based rapid detection test for the diagnosis of SARS-CoV-2 in respiratory samples. Int J Infect Dis. 2020;99:328-33.
Scohy A, Anantharajah A, Bodéus M, Kabamba-Mukadi B, Verroken A, Rodriguez-Villalobos H. Low performance of rapid antigen detection test as frontline testing for COVID-19 diagnosis. J Clin Virol. 2020;129:104455.
Lambert-Niclot S, Cuffel A, Le Pape S, Vauloup-Fellous C, Morand-Joubert L, Roque-Afonso AM, et al. Evaluation of a rapid diagnostic assay for detection of SARS-CoV-2 antigen in nasopharyngeal swabs. J Clin Microbiol. 2020;58(8):10.1128/jcm.00977-20.
Brümmer LE, Katzenschlager S, Gaeddert M, Erdmann C, Schmitz S, Bota M, et al. Accuracy of novel antigen rapid diagnostics for SARS-CoV-2: A living systematic review and meta-analysis. PLoS Med. 2021;18(8):e1003735.
Carpenter CR. Rapid antigen and molecular tests had varied sensitivity and ≥97% specificity for detecting SARS-CoV-2 infection. Ann Intern Med. 2020;173(12):JC69.
Department of Disease Control. Guidelines for ATK Testing in factories and slum communities [Internet]. 2021 [cited 2021 Dec 25]. Available from: https://ddc.moph.go.th/viralpneumonia/file/g_srrt/g_srrt_221264.pdf
Moslehi S, Rabiei N, Soltanian AR, Mamani M. Application of machine learning models based on decision trees in classifying the factors affecting mortality of COVID-19 patients in Hamadan, Iran. BMC Med Inform Decis Mak. 2022;22(1):192.
Chaimayo C, Kaewnaphan B, Tanlieng N, Athipanyasilp N, Sirijatuphat R, Chayakulkeeree M, et al. Rapid SARS-CoV-2 antigen detection assay in comparison with real-time RT-PCR assay for laboratory diagnosis of COVID-19 in Thailand. Virol J. 2020;17:1-7.
Amer RM, Samir M, Gaber OA, El-Deeb NA, Abdel Moaty AA, Ahmed AA, et al. Diagnostic performance of rapid antigen test for COVID-19 and the effect of viral load, sampling time, subject’s clinical and laboratory parameters on test accuracy. J Infect Public Health. 2021;14(10):1446-53.
Hase R, Kurita T, Mito H, Yano Y, Watari T, Otsuka Y, et al. Potential for false positive results with quantitative antigen tests for SARS-CoV-2: a case of a child with acute respiratory infection. J Infect Chemother. 2022;28(2):319-20.
Zoabi Y, Deri-Rozov S, Shomron N. Machine learning-based prediction of COVID-19 diagnosis based on symptoms. NPJ Digit Med. 2021;4(1):3.
Nicolau I, Pérez-Gómez M, Paredes D, Trinh NTH, Roel E, Burn E. Association of COVID-19 vaccination with respiratory disease severity in patients with COVID-19: a cohort study. Lancet Respir Med. 2023;11(10):825-834.
Ahmad A, Safi O, Malebary S, Alesawi S, Alkayal E. Decision tree ensembles to predict coronavirus disease 2019 infection: a comparative study. Complexity. 2021;2021:1-8.
Pal M, Parija S, Mohapatra RK, Mishra S, Rabaan AA, Al Mutair A, et al. Symptom-based COVID-19 prognosis through AI-based IoT: a bioinformatics approach. Biomed Res Int. 2022;2022.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2024 University of Phayao
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
ผู้นิพนธ์ต้องรับผิดชอบข้อความในบทนิพนธ์ของตน มหาวิทยาลัยพะเยาไม่จำเป็นต้องเห็นด้วยกับบทความที่ตีพิมพ์เสมอไป ผู้สนใจสามารถคัดลอก และนำไปใช้ได้ แต่จะต้องขออนุมัติเจ้าของ และได้รับการอนุมัติเป็นลายลักษณ์อักษรก่อน พร้อมกับมีการอ้างอิงและกล่าวคำขอบคุณให้ถูกต้องด้วย
The authors are themselves responsible for their contents. Signed articles may not always reflect the opinion of University of Phayao. The articles can be reproduced and reprinted, provided that permission is given by the authors and acknowledgement must be given.
