Detection of Chilling Injury in Cucumis Sativus Using Bioimpedance Spectroscopy and Machine Learning Algorithms

This paper introduces an accessible, and cost-effective bioimpedance spectroscopy (BIS) system designed for the early detection of chilling injury in Cucumis sativus (cucumber). This integrated hardware and software solution aims to advance beyond traditional visual and statistical methods by offering real-time and reliable analysis. Enhanced with a specialized electrode configuration, the system utilizes advanced machine learning (ML) techniques, including Principal Component Analysis (PCA) and t-Distributed Stochastic Neighbor Embedding (t-SNE), to analyze and refine bioimpedance data efficiently. Comparative methods reveal that the combination of t-SNE and Gradient Boosted Decision Trees (GBDT) outperforms other methods, achieving an accuracy of 92.5%, precision of 90.5%, recall of 95.0%, and F1-Score of 92.7%. These results underscore the effectiveness of the developed BIS system in non-invasively detecting morphological changes in plant attributes due to chilling injuries. The proposed approach holds promise for revolutionizing postharvest quality management of cucumbers and has the potential to be adapted for a wider range of agricultural products.