Integrating Machine Learning with Flow-Imaging Microscopy for Automated Monitoring of Algal Blooms

Abstract

Real-time monitoring of phytoplankton in freshwater systems is critical for early detection of harmful algal blooms so as to enable efficient response by water management agencies. This paper presents an image processing pipeline developed to adapt ARTiMiS, a low-cost automated flow-imaging device, for real-time algal monitoring specifically in freshwater and environmental systems. This pipeline addresses several challenges associated with autonomous imaging of aquatic samples such as flow-imaging artifacts (i.e., out-of-focus and background objects), as well as specific challenges associated with monitoring of open environmental systems (i.e., identification of novel objects). The pipeline leverages a Random Forest model to identify out-of-focus particles with an accuracy of 89% and a custom background particle detection algorithm to identify and remove particles that erroneously appear in consecutive images with >97±2.8% accuracy. Furthermore, a convolutional neural network (CNN), trained to classify distinct classes comprising both taxonomical and morphological categories, achieved 94% accuracy in a closed dataset. Nonetheless, the supervised closed-set classifiers struggled with the accurate classification of objects when challenged with debris and novel particles which are common in complex open environments; this limits real-time monitoring applications by requiring extensive manual oversight. To mitigate this, three methods incorporating classification with rejection were tested to improve model precision by excluding irrelevant or unknown classes. Combined, these advances present a fully integrated, end-to-end solution for real-time HAB monitoring in open environmental systems thus enhancing the scalability of automated detection in dynamic aquatic environments.