Particle Counting Method To Improve Sensitivity of Paper-Based Microfluidic Applications Using Smartphone-Based Detection

Abstract

With the increase in demand for rapid biosensor devices, the development of a more sensitive and cost-efficient biosensor device or technique becomes more needed. In this thesis, the development of a particle counting method using a smartphone-based platform and a paper-based microfluidic chip for the application of a competitive inhibition immunoassay is discussed. In the two chapters of this thesis, the detection of α-amanitin (α-AMA) and (-)-trans-delta-tetrahydrocannabinol (THC) is discussed. The reason the two targets will be discussed is due to the similarity in their small molecular size which is ideal for the use of a competitive inhibition immunoassay. Firstly, a paper-based microfluidic competitive immunoassay using a smartphone-based fluorescence microscope for the detection of α-AMA in toxic mushrooms was developed. The low limit of detection (LOD) (1 pg/mL) of the competitive immunoassay due to the particle counting methodology allowed the utilization of a rinsing method over the traditional extraction method, which is more time consuming and requires a centrifuge to perform. Secondly, a competitive immunoassay was also developed for the detection of THC in clinical saliva samples. The LOD when using this method is 0.5 pg/mL. Using a support vector machine (SVM), a machine learning algorithm for classification, THC concentration was able to be predicted with an accuracy of over 90% for one saliva sample and quantified. In conclusion, the use of the particle counting method used in both applications has shown lower LOD than other attempts at detecting the same targets described and the overall work process is cost-effective and friendly for fieldwork applications.