TY - GEN
T1 - MACHINE LEARNING-BASED QUANTITATIVE ANALYSIS METHOD USING IMMUNO-WALL DEVICE
AU - Shin, Jungchan
AU - Kasama, Toshihiro
AU - Miyake, Ryo
N1 - Publisher Copyright:
© 2022 MicroTAS 2022 - 26th International Conference on Miniaturized Systems for Chemistry and Life Sciences. All rights reserved.
PY - 2022
Y1 - 2022
N2 - We have newly designed antigen concentration estimation method utilizing machine learning technique on the immuno-wall device. Fluorescence intensities of one antigen detected under different amounts of capture antibody were combined into one combination, so that it can generate one unique pattern which is dependent on concentration of antigen (gradation immunoassay). In this research, 46,656 combinations were prepared by using 6 different concentrations of antigen and 5 different concentrations of capture antibody. Multivariate linear regression was conducted to estimate concentration of antigen using fluorescence intensities obtained under different amount of capture antibody as variables. As a result, comparison of the predictions to the actual antigen concentrations were drawn, and the coefficient of determination was 0.986. This method can provide accurate immunoassay result at on-site diagnosis.
AB - We have newly designed antigen concentration estimation method utilizing machine learning technique on the immuno-wall device. Fluorescence intensities of one antigen detected under different amounts of capture antibody were combined into one combination, so that it can generate one unique pattern which is dependent on concentration of antigen (gradation immunoassay). In this research, 46,656 combinations were prepared by using 6 different concentrations of antigen and 5 different concentrations of capture antibody. Multivariate linear regression was conducted to estimate concentration of antigen using fluorescence intensities obtained under different amount of capture antibody as variables. As a result, comparison of the predictions to the actual antigen concentrations were drawn, and the coefficient of determination was 0.986. This method can provide accurate immunoassay result at on-site diagnosis.
KW - Antigen concentration estimation
KW - Immuno-wall device
KW - Machine learning
UR - https://www.scopus.com/pages/publications/85175733616
UR - https://www.scopus.com/pages/publications/85175733616#tab=citedBy
M3 - Conference contribution
AN - SCOPUS:85175733616
T3 - MicroTAS 2022 - 26th International Conference on Miniaturized Systems for Chemistry and Life Sciences
SP - 1161
EP - 1162
BT - MicroTAS 2022 - 26th International Conference on Miniaturized Systems for Chemistry and Life Sciences
PB - Chemical and Biological Microsystems Society
T2 - 26th International Conference on Miniaturized Systems for Chemistry and Life Sciences, MicroTAS 2022
Y2 - 23 October 2022 through 27 October 2022
ER -