Machine learning assisted and smartphone integrated homogeneous electrochemiluminescence biosensor platform for sample to answer detection of various human metabolites.

The sensitive and accurate detection of glucose and lactate is essential for early diagnosis and effective management of diabetes complications. Herein, a 3D Printed ECL imaging system integrated with a Smartphone has been demonstrated to advance the traditional ECL to make a portable, affordable, and turnkey point-of-care solution to detect various human metabolites. A universal cross-platform application was introduced for analyzing ECL emitted signals to automate the whole detection process for real-time monitoring and rapid diagnostics. The developed ECL system was successfully applied and validated for detecting glucose and lactate using a single-electrode ECL biosensing platform. For glucose and lactate detection, the device showed a linear range from 0.1 mM to 1 mM and 0.1 mM-4 mM with a detection limit (LoD) of 0.04 mM and 0.1 mM, and a quantification limit (LoQ) of 0.142 mM and 0.342 mM, respectively. The developed method was evaluated for device stability, accuracy, interference, and real sample analysis. Furthermore, to assist in selecting the accurate and economic ECL sensing platform, SE-ECL devices fabricated via different fabrication approaches such as Laser-Induced Graphene, Screen Printing, and 3D Printing are studied for the conductivity of electrode and its significance on ECL signal. It was observed that emitted ECL signal is independent of the electrical conductivity for the same concentration of analytes. The findings suggested that the developed miniaturized point-of-care ECL platform would be a comprehensive and integrated solution for detecting other human metabolites and have the potential to be used in clinical applications.