WASIMM

WASIMM project aims to develop and implement innovative technologies and methodologies such as total internal reflection fluorescence (TIRF) microscopy and machine learning (ML) to achieve rapid and highly sensitive detection of coliform bacteria and viruses in wastewater, thereby enhancing water safety and public health protection in compliance with revised stringent regulations. The key objective is advancing water detection methodologies, to detect specific viruses and bacteria rapidly and efficiently, in-line with the water chain, for timely preventing outbreaks. This will be achieved via research, development, and lab validations in controlled environment of advanced methodologies targeting to detect and identify viruses and bacteria within large volumes of wastewater. Novel preconcentration technologies already being developed by CyRIC are planned to be combined in the future with the WASIMM detection system for the first time in an automated system.

Innovation beyond the state-of-the-art
Currently used biological analytical methods for microorganism detection, are complicated lab-based, requiring manual time-consuming pretreatment steps such as DNA extraction. WASIMM aims to combine TIRF microscopy and ML to surpass these methods and address these issues providing real-time, label-free detection system, reducing the time and expertise required for analysis while maintaining high accuracy. The system will reach the technological readiness level 4 (TRL4-via study, design, development, and lab testing). The combination of the TIRF microscopy high-resolution imaging capabilities with the analytical power of ML algorithms offers several advantages such as (a) enhanced sensitivity leading to individual bacterial cell and virus detection, enabling the identification of low concentrations of coliforms that may pose health risks but could be missed by conventional methods, (b) real-time analysis with ML algorithms rapidly processing TIRF-generated images, enabling automated and real-time detection of coliforms, reducing the time required for analysis compared to culture-based methods, (c) accuracy and reliability by training ML models with diverse datasets, improving their ability to accurately identify coliform bacteria amidst variations in environmental conditions and bacterial morphologies, (d) cost and labour efficiency due to the automated nature of ML-based analysis reducing the need for extensive manual labour, ultimately decreasing operational costs in the long term. The proposed project serves as a foundation for the development of automated portable detection devices, potentially enabling on-site, in-line, rapid detection in various settings like wastewater quality monitoring or even clinical diagnostics.