Machine Learning-Assisted Automatically Electrochemical Addressable Cytosensing Arrays for Anticancer Drug Screening

Abstract

The high-throughput and accurate screening of anticancer drugs is crucial to the preclinical assessment of candidate drugs and remains challenging. Herein, an automatically electrochemical addressable cytosensor (AEAC) for the efficient screening of anticancer drugs is reported. This sensor consists of sectionalized laser-induced graphene arrays decorated by the rhombohedral TiO2 and spherical Pt nanoparticles (LIG-TiO2–Pt) with high electrocatalytic activity for H2O2 and a homemade Ag/Pt electrode couple fixed onto the robot arm. The immobilization of laminin on the surface of LIG-TiO2–Pt can promote its biocompatibility for the growth and proliferation of various tumor cells, which empowers the in situ monitoring of H2O2 directly released from these live cells for drug screening. A machine learning (ML) algorithm is employed to eliminate the possible random or systematic errors of AEAC, realizing rapid, high-throughput, and accurate prediction of different types of anticancer drugs. This ML-assisted AEAC provides a powerful approach to accelerate the evolution of sensing-served tumor therapy.