Evaluating the Performance of Ethanol Electrochemical Nanobiosensor Through Machine for Predictive Analysis of Electric Current in Self-Powered Biosensors

In this study, the focus is on ethanol nano biosensors based on alcohol oxidase (AOX) enzymatic reactions and the feasibility of generating electric current for biobatteries. The aim is to convert the latent energy in ethanol into electrical energy through the enzymatic oxidation process in the presence of an AOX enzyme. The release of electrons and the creation of a potential difference make the use of ethanol as a biofuel cell/self-power biosensor in biologically sensitive systems feasible. To achieve this, glassy carbon electrodes were modified with gold nanoparticles to enhance conductivity, and the AOX enzyme was immobilized on the working electrode. The current generated through the enzymatic process was measured in various pH and analyte concentration conditions. Afterward, machine-learning models, including multilayer perceptron (MLP), deep neural network, decision tree, and random forest, were employed to assess the impact of parameters on electric current generation, evaluate the error rate, and compare the results. The results indicated that the MLP model was the most suitable method for predicting the electric current produced under different pH, temperature, and ethanol concentration values. These findings can be utilized to identify optimal conditions and increase the current output for use as a reliable energy source in self-powered biosensors. In conclusion, this study suggests a promising way to generate electricity by oxidizing ethanol with the AOX enzyme. The use of machine learning to analyze experimental data has provided insight into optimal conditions for maximizing electric current output for developing sustainable energy sources in biologically sensitive systems and biobattery technology.