Modular GrowBox Design for Precision Monitoring of Plant Bioelectrical Signals

Grow boxes are controlled plant growth chambers that could provide standardized environments for experiments in plant electrophysiology. This paper proposes a grow box system with integrated sensors, actuators, and an 8-channel biosignal amplifier. This allows for precise, real-time monitoring, and manipulation of growth conditions. We demonstrate its practical application through experiments isolating light-induced action potentials (APs) in plants. Then, we evaluate two electrode configurations to identify the optimal setup for signal acquisition. Additionally, we implement supervised machine learning, using the Extreme Gradient Boosting (XGBoost) algorithm, to automatically classify plant physiological states. The model achieves a 96.8% accuracy in detecting changes associated with nycthemeral rhythms. These results highlight the potential of customizable grow boxes, like the system presented, combined with standard-ized recording protocols to facilitate seamless data sharing and collaborative efforts for refining analytical techniques in plant electrophysiology.