Search

Global food production needs to increase in order to meet the demands of an ever growing global population. As resources are finite, the most feasible way to meet this demand is to minimize losses and improve efficiency. Regular monitoring of factors like animal health, soil and water quality for example, can ensure that the resources are being used to their maximum efficiency. Existing monitoring techniques however have limitations, such as portability, turnaround time and requirement for additional reagents. In this work, we explore the use of micro- and nano-scale electrode devices, for the development of an electrochemical sensing platform to digitalize a wide range of applications within the agri-food sector. With this platform, we demonstrate the direct electrochemical detection of pesticides, specifically clothianidin and imidacloprid, with detection limits of 0.22 ng/mL and 2.14 ng/mL respectively, and nitrates with a detection limit of 0.2 µM. In addition, interdigitated electrode structures also enable an in-situ pH control technique to mitigate pH as an interference and modify analyte response. This technique is applied to the analysis of monochloramine, a common water disinfectant. Concerning biosensing, the sensors are modified with bio-molecular probes for the detection of both bovine viral diarrhea virus species and antibodies, over a range of 1 ng/mL to 10 µg/mL. Finally, a portable analogue front end electronic reader is developed to allow portable sensing, with control and readout undertaken using a smart phone application. Finally, the sensor chip platform is integrated with these electronics to provide a fully functional end-to-end smart sensor system compatible with emerging Agri-Food digital decision support tools.

Sensor technologies that measure grazing and ruminating behaviour as well as physical activities of individual cows are intended to be included in precision pasture management. One of the advantages of sensor data is they can be analysed to support farmers in many decision-making processes. This article thus considers the performance of a set of RumiWatchSystem recorded variables in the prediction of insufficient herbage allowance for spring calving dairy cows. Several commonly used models in machine learning (ML) were applied to the binary classification problem, i.e., sufficient or insufficient herbage allowance, and the predictive performance was compared based on the classification evaluation metrics. Most of the ML models and generalised linear model (GLM) performed similarly in leave-out-one-animal (LOOA) approach to validation studies. However, cross validation (CV) studies, where a portion of features in the test and training data resulted from the same cows, revealed that support vector machine (SVM), random forest (RF) and extreme gradient boosting (XGBoost) performed relatively better than other candidate models. In general, these ML models attained 88% AUC (area under receiver operating characteristic curve) and around 80% sensitivity, specificity, accuracy, precision and F-score. This study further identified that number of rumination chews per day and grazing bites per minute were the most important predictors and examined the marginal effects of the variables on model prediction towards a decision support system.