A machine-learning-integrated portable electrochemiluminescence sensing platform for the visualization and...

Electrochemiluminescence (ECL)-based point-of-care testing (POCT) has the potential to facilitate the rapid identification of diseases, offering advantages such as high sensitivity, strong selectivity, and minimal background interference. However, as the throughput of these devices increases, the issues of increased energy consumption and cross-contamination of samples remain. In this study, a high-throughput ECL biosensor platform with the assistance of machine learning algorithms is developed by combining a microcolumn array electrode, a microelectrochemical workstation, and a smartphone with custom software. The microcolumn array electrode is modified with gold nanoparticles by the electrodeposition method to enhance the electrical conductivity and effectively catalyze the luminescence reaction, leading to a significantly enhanced ECL intensity. The support vector machine (SVM) algorithm is employed to analyze the signals from luminescent images captured by the smartphone, enabling the quantitative detection of the SARS-CoV-2 nucleocapsid (SARS-CoV-2 N) protein with a linear detection range from 0.001 to 10 ng/mL and a limit of detection as low as 0.86 pg/mL. The application of the SVM model and a backpropagation (BP) neural network algorithm, both leveraging RGB feature extraction, has demonstrated the capability to effectively classify and predict the concentration of the target protein with high accuracy. This machine learning-assisted ECL-POCT platform significantly reduces cross-contamination and signal interference in traditional high-throughput ECL systems, providing great potential for large-scale and simultaneous disease screening.