Open AccessDetection Limit of the Four-Parameter Logistic Model for the Quantitative Detection of Serum Squamous Cell Carcinoma Antigenin Cervical Cancer Based on Surface Plasmon Resonance Biosensor
Author(s) -
Noor Azlina Masdor
Publication year - 2021
Publication title -
journal of environmental microbiology and toxicology
Language(s) - English
Resource type - Journals
ISSN - 2289-5906
DOI - 10.54987/jemat.v9i2.644
Subject(s) - sigmoid function , calibration curve , confidence interval , surface plasmon resonance , linearization , calibration , logistic regression , mathematics , statistics , mathematical analysis , nonlinear system , detection limit , physics , artificial intelligence , computer science , materials science , nanotechnology , nanoparticle , quantum mechanics , artificial neural network
Biochemical diagnostic procedures, such as protein binding, rely on biomolecular interactions as its diagnostic modality, and as a consequence, their calibration curves are more complex. In addition, sigmoidal curves are often seen in these tests. In the event of asymmetry, a logistic (5PL) curve, or a logistic (4PL) curve, may be the best way to represent a distinctive sigmoidal relationship. It is possible that the linearization of an otherwise nonlinear connection by log transformation may result in a disruption of the error structure of the curve, and that this will have the opposite effect of decreasing or even eliminating error in the relationship. Previously, a surface plasmon resonance biosensor for the detection of squamous cell carcinoma antigen (SCCa) using nanoparticle technology was developed. However, based on the calibration curve, it conforms to the majority on sigmoidal shape curve for antibody-type sensing system. The resultant curve showed a sigmoidal calibration curve but was not modelled according to any of the sigmoidal models available. The LOD value obtained through the 4PL modelling exercise based on the classical method was 0.255 pM (95% confidence interval of 0.167 to 0.379) while the pooled standard deviation (PSD) method yielded an LOD value of 0.035 pM (95% confidence interval of 0.011 to 0.067), which indicates that the PSD method was superior.