https://doi.org/10.1140/epjp/s13360-023-03988-1
Regular Article
Taguchi method: artificial neural network approach for the optimization of high-efficiency microfluidic biosensor for COVID-19
1
Laboratory of Electronics and Microelectronics, Faculty of Science of Monastir, University of Monastir, 5019, Monastir, Tunisia
2
Quantum and Statistical Physics Laboratory, Faculty of Sciences of Monastir, University of Monastir, 5019, Monastir, Tunisia
3
Higher National Engineering School of Tunis, Taha Hussein Montfleury Boulevard, University of Tunis, 1008, Tunis, Tunisia
4
Higher Institute of Applied Sciences and Technology of Sousse, Sousse, Tunisia
5
Department of Physics, College of Science, Princess Nourah Bint Abdulrahman University, Riyadh, Saudi Arabia
6
Department of Physics, College of Science, Majmaah University, 11952, Al Majma’ah, Saudi Arabia
Received:
27
February
2023
Accepted:
12
April
2023
Published online:
27
April
2023
COVID-19 is a pandemic disease caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). This virus is mainly spread by droplets, respiratory secretions, and direct contact. Caused by the huge spread of the COVID-19 epidemic, research is focused on the study of biosensors as it presents a rapid solution for reducing incidents and fatality rates. In this paper, a microchip flow confinement method for the rapid transport of small sample volumes to sensor surfaces is optimized in terms of the confinement coefficient β, the position of the confinement flow X, and its inclination α relative to the main channel. A numerical simulation based on two-dimensional Navier–Stokes equations has been used. Taguchi’s L9(33) orthogonal array was adopted to design the numerical assays taking into account the confining flow parameters (α, β, and X) on the response time of microfluidic biosensors. Analyzing the signal-to-noise ratio allowed us to determine the most effective combinations of control parameters for reducing the response time. The contribution of the control factors to the detection time was determined via analysis of variance (ANOVA). Numerical predictive models using multiple linear regression (MLR) and an artificial neural network (ANN) were developed to accurately predict microfluidic biosensor response time. This study concludes that the best combination of control factors is that corresponds to
,
and X = 40 µm. Analysis of variance (ANOVA) shows that the position of the confinement channel (62% contribution) is the factor most responsible for the reduction in response time. Based on the correlation coefficient (R2), and value adjustment factor (VAF), the ANN model performed better than the MLR model in terms of prediction accuracy.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2023. Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.