Real-Time Estimation of Corneal Damage Using IL-6 in Tears Based on Flexi-Graphene FET Sensor Array

Increased levels of IL-6 in tear films have been associated with progressive severity of corneal damage in patients. However, most of the commercially approved method of marker detection in tears either involve lateral flow immunochromatography assay or enzyme linked immunosorbent assay (ELISA), which suffer from low sensitivity and operational complexities, respectively. Recent efforts reported for affordable, sensitive, and reliable detection of tear markers using ultrasensitive field-effect transistor (FET) biosensor involve multiple washing steps to reduce the stochastic fluctuations in the measurement arising from the nonspecific proteins in tears. Thus, developing a convenient point-of-care (PoC) platform for sensitive real-time quantification of IL-6 in tears is an urgent need for rapid estimation of the severity of corneal damage, leading to targeted therapeutic treatment. In this letter, we demonstrate real-time estimation of IL-6 in tears for the first time using thermally reduced graphene oxide FETs on flexible polycarbonate substrate. It has been observed that for real-time detection of IL-6 in tears, without involving any intermediate washing steps, the sensor response results in nonseparable steady-state outputs leading to inaccurate classification and quantification. This challenge has been mitigated by extracting multiple features, such as Dirac point voltage, percentage change in drain–source current, and maximum and minimum transconductance from the sensor characteristics, and subjecting them to random forest-based machine learning algorithm. The proposed methodology realizes a real-time clinically sensitive detection limit of 1 pg/ml IL-6 in tears with a detection time of 10 min, thereby paving the way toward an effective PoC platform for estimation of tear biomarkers.