An IoT-based smart biosensor for the measurement of nitrate concentration in liquid samples

Various enzymatic bioanalytical sensors have been developed for the specific and sensitive detection of nitrate on the basis of enzymatic nitrate to nitrite reduction. Since the enzyme immobilized on the surface of the electrode of the biosensor is a living organism and its activity reduces over time, it needs to be replaced frequently, limiting its application as a portable device. Moreover, embedding a data storage element and processor for decision-making based on machine learning in the structure of the device to consider the enzyme activity during the nitrate estimation increases its costs, undesirable for mass production. In this work, a cost-effective smart nitrate biosensor is introduced that benefits from an IoT platform. The electrochemical characteristics of the samples along with the enzyme’s lifespan are transmitted to a server where a support vector machine with hyperparameters optimized by the slime mold algorithm predicts the nitrate concentration in the samples, followed by sharing the results with the clients. The results showed that the machine learning unit is capable of predicting the nitrate concentration with R
²
and MSE of 0.92 and 0.036, respectively, three weeks after immobilizing the nitrate reductase on the surface of the biosensor’s electrode. Besides, the IoT platform developed in this study can reduce the cost of the smart portable nitrate biosensors by ca. 25% due to considering a unique decision-making unit in the server instead of considering a processor capable of implementing machine learning tasks for each sensor. This can also be useful to gather all information from the nitrate sensors for environmental pollution monitoring on a large scale.