Monitoring the quality of recreational waters such as beaches and rivers is becoming a global concern to protect human health (e.g. the EU Bathing Water Directive (BWD) in Europe). In Scotland, water quality at 86 designated bathing water sites is assessed by the Scottish Environment Protection Agency (SEPA) from May to September. However, methods for detecting pathogens such as E-coli are challenging as samples need to be collected from several locations and transported to the laboratory within a short period of time, typically within 6 hours. Therefore, authorities responsible for these bathing waters would ideally require new tools to continually monitor levels of pathogens on-site. The aim of this poster is to present the findings from initial investigations into the development of a microfluidic system for the processing of water samples including a new approach to improve the quality of images from a microscopic camera. This will then allow image processing and a deep learning algorithm to be investigated for the detection and classification of microorganisms. The system will be tested on real water samples with collaborators at the SEPA. This is a multi-disciplinary project involving School of Engineering and the Built Environment, School of Computing, School of Applied Sciences, and SEPA. It is funded by The Data Lab and SEPA.