Machine learning enabled onsite electrochemical detection of lidocaine using a microneedle array integrated screen printed electrode

Despite several advantageous uses of lidocaine patches to overcome discomfort and pain in various clinical settings, overdosage of this drug can cause unwanted side effects on the cardiovascular and central nervous system, which can lead to life-threatening conditions. Therefore, the development of a rapid, sensitive, and userfriendly point-of-care device for onsite lidocaine detection is of great clinical importance. To address this issue, we have developed a machine learning enabled wireless microneedle array integrated screen-printed electrodebased electrochemical point-of-care device for rapid and effective detection of lidocaine. The fabricated device utilizes novel ultra-sharp microneedles arrays having a reservoir in its base, which are designed to collect the interstitial fluid through open side channels, and graphene-modified screen-printed carbon electrodes for the electrochemical detection of lidocaine. Under optimal conditions, the developed sensor exhibited high sensitivity and good selectivity towards lidocaine along with a linear current response over the detection range from 1-120 mu M with the lowest detection limit of 0.13 mu M. In addition, to make the device user friendly, a machine learning model was developed using experimental sensing data to predict the lidocaine concentration and further deployed to prepare a web application for digital visualization of lidocaine concentration.