Aquatic ecosystems are vital in regulating climate and providing resources, but face threats from globalchange and local stressors. Understanding their dynamics is crucial for sustainable use and conservation.Coordinated by the EGI Foundation, which leverages a Federation of over 200 Data Centres all overEurope and delivered over 82M Cloud CPU/hours to more than 95,000 users, the iMagine project offersaquatic science researchers (marine and freshwaters) the iMagine AI Platform, a suite of AI-poweredimage analysis tools for researchers in aquatic sciences, facilitating a better understanding of scientificphenomena and applying AI and ML for processing image data. The platform supports the entiremachine learning cycle, from model development to deployment, leveraging data from underwaterplatforms, webcams, microscopes, drones, and satellites and utilising distributed resources acrossEurope. A serverless architecture and DevOps approach enable easy sharing and deployment of AImodels. Four providers within the pan-European EGI federation power the platform, offering substantialcomputational resources for image processing.Eight use cases in iMagine focus on image analytics services, which will be available to externalresearchers through Virtual Access. FlowCam, powered by VLIZ, is one of them.Phytoplankton, the single-cell algae at the basis of marine food webs and an essential indicator ofecosystem health, is continuously being monitored at several stations in the Belgian Part of the NorthSea under the LifeWatch research infrastructure. To process monitoring samples in a fast andautomated manner, we make use of automated imaging techniques like FlowCam. By aligning particlesin the sample in a continuous fluid stream and capturing each particle in a picture as it passes a camera,this device can produce an image library of a sample in under 30 minutes. While FlowCam hassignificantly sped up time spent in the lab, it delivers about 350,000 images yearly and calls for anautomated approach to handle high data loads. To speed up the taxonomist’s job of manually labellingall these images, we built semi-automated data pipelines and implemented machine-learningalgorithms, specifically Convolutional Neural Networks, to classify the images. Over the years, thiscombination of automated imaging and machine learning has helped us build a set of over 2,2 millionannotated FlowCam images and trained classifiers fine-tuned by taxonomists correcting wrong modelpredictions. This dataset and the trained classifiers have proven to greatly benefit our marinemonitoring, and we wanted to share this asset with other researchers. Under iMagine, we aim to publish the open access image set and classifiers and build a user-friendlymodule where users can both predict FlowCam images using pretrained models and train classifiers ontheir own image input. The iMagine platform hosting this module offers an integrated environment withall source code and a graphical user interface for users with less coding experience. Computingresources for the services are also available to the user through the platform. The FlowCam modulefurther provides tools for post-hoc analysis of model performance and code for image transformationand augmentation to deal with different image resolutions and class imbalances in training sets. Moreinformation on the project and the FlowCam service can be found at https://www.imagine-ai.eu. In thenext coming years, we hope to facilitate many marine researchers in the application of automatedclassification of phytoplankton imaging data. We actively encourage researchers and monitoringprograms to make use of the FlowCam service and the iMagine platform to contribute to more efficientbiomonitoring. The outcomes of the FlowCam use case could boost the awareness of the Belgian openscience community about the healthiness of the national marine ecosystem through the projectfindings, which could in turn be beneficial to researchers in marine science all over Europe.