Machine learning enabled onsite electrochemical detection of lidocaine using a microneedle array integrated screen printed electrode
Machine learning enabled onsite electrochemical detection of lidocaine using a microneedle array integrated screen printed electrode
复制标题
DOI：
10.1016/j.electacta.2023.143664
复制
发表时间：
2024-01
影响因子：
6.6
通讯作者：
Sachin Kadian;S. Sahoo;Pratima Kumari;Roger J. Narayan
中科院分区：
文献类型：
--
作者：
Sachin Kadian;S. Sahoo;Pratima Kumari;Roger J. Narayan
尽管利多卡因贴片在各种临床环境中有几种有利的用途来克服不适和疼痛，但过量使用这种药物会对心血管和中枢神经系统产生意想不到的副作用，这可能导致危及生命的疾病。因此，开发一种快速、灵敏、用户友好的现场检测利多卡因设备具有重要的临床意义。为了解决这个问题，我们开发了一种机器学习无线微针阵列集成丝网印刷电极电化学护理点设备，用于快速有效地检测利多卡因。该装置采用新型的超锋利微针阵列，其底部有一个储液池，设计用于通过开放的侧通道收集间隙液，石墨烯修饰的丝网印刷碳电极用于利多卡因的电化学检测。在最佳条件下，该传感器对利多卡因具有较高的灵敏度和良好的选择性，在1 ~ 120µM的检测范围内具有线性电流响应，最低检出限为0.13µM。此外，为了使设备用户友好，我们利用实验传感数据开发了一个机器学习模型来预测利多卡因浓度，并进一步部署了一个web应用程序来实现利多卡因浓度的数字可视化。
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 user-friendly 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 electrode-based 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 µM with the lowest detection limit of 0.13 µM. 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.