Evaluation of Neural Network for Automated Classification of Plant Component on Histological Section

Classification of plant component on image histological sections is critical for determining non-compliance type of undeclared additiveand further action for technologist, or other responsible person. However, this task is often challenging due to the absence of professional histologists or non-compliance with the conditions of microstructural analysis and also the subjective criteria for evaluation. In this study, we propose a machine learning model that automatically classifies the plant component on images histological sections. Our model uses a convolutional neural network to identify regions of plant components, then aggregates those classifications to infer predominant and minor plant components on histological sections image. We evaluated our model on an independent set of 95 images histological sections. It achieved a kappa score of 0.525 and an agreement of 66.6% with three histologists for classifying the predominant plant component, slightly higher than the inter-histologists kappa score of 0.485 and agreement of 62.7% on this test set. All evaluation metrics for our model and the three histologists were within 95% confidence intervals of agreement.