A Machine Learning-Assisted Highly Sensitive Photoelectrochemical Biosensor Based on Dual-Signal CuO Nanoparticles for the Combined Detection of Dual LncRNAs

As an emerging biomarker of lung cancer, developing a highly sensitive method for the combined detection of dual lncRNAs holds significant value for the early screening of lung cancer. Herein, a novel photoelectrochemical (PEC) biosensor based on dual-signal copper oxide nanoparticles is developed for the combined detection of dual lncRNAs, utilizing machine learning to enhance the intelligent screening efficiency of lung cancer. The as-synthesized copper oxide NPs exhibit a high photoelectric conversion efficiency and dual-photocurrent signals. Through lncRNA-triggered catalytic hairpin assembly signal amplification, abundant tungsten disulfide quantum dots are introduced into the photoelectrode to reduce the photocurrent response. The proposed PEC biosensing platform has high sensitivity with low detection limits of 72 aM for HOTAIR and 78 aM for MALAT1. Furthermore, this PEC platform for dual lncRNAs in whole blood can successfully distinguish cancer patients from healthy individuals. Moreover, machine learning is employed to explore the hidden potential patterns in the established PEC method, and the accuracy rate of intelligent screening for lung cancer reaches 93%. The integrated detection system of machine learning and PEC biosensors offers a new method for dual lncRNA analysis and cancer early screening.