b'AI-Driven Biofilm Detectionfor Sustainable Fermentation CASE STUDYBiofilm formation during industrial fermentationthis supports public goals of sustainable food presents a significant challenge for foodproduction and climate change mitigation. Professor production companies like Quorn Foods. BiofilmsAnnalisa Occhipinti from Teesside University said, accumulate on fermentor walls, causing lumps in the final product and often leading to prematureNBICs support enabled access to valuable industrial termination of the fermentation cycle. This resultsdatasets and facilitated close collaboration between in productivity loss, higher production costs, andQuorn fermentation experts (Mark Taylor) and AI increased energy waste, impeding the sectors goalresearchers at Teesside University. This led to the of sustainable, low-carbon food production. development of a successful semantic segmentation pipeline based on a UNet architecture that can An NBIC-funded Flexible Talent Mobilitydetect biofilm presence with high accuracy. The Accounts (FTMA) project between researchersNBIC funding helped overcome data annotation at Teesside University and Quorn Foodschallenges and supported iterative model developed an AI-driven biofilm detection systemimprovements by providing industry feedback.using microscopy images collected at different stages of Quorns fermentation process.Following the NBIC collaboration, the team successfully secured BBSRC funds to expand this By leveraging transfer learning on Convolutionalresearch. This funding, led by Professor Claudio Neural Networks and integrating meta- Angione at Teesside University, supports further information via multi-modal approaches, thedevelopment and scaling of the AI model, accelerating model semantically segmented biofilm regionsits translation into industrial applications. to predict biofilm percentage. This enabled the optimisation of fermentation cycle termination timing to reduce waste and energy consumption. The AI model can identify biofilm formation in microscopy images and predict optimal fermentation termination times, directly addressing quality control and reducing process inefficiencies. This is expected to lower energy use and carbon emissions, aligning with Clean Growth objectives. Quantitative validation and deployment are ongoing.By improving fermentation process control, the technology benefits the food production industry through enhanced product quality, reduced waste, and a lower environmental footprint. Indirectly, Professor Annalisa Occhipinti, Teesside University Professor Occhipinti develops multimodal AI integrating computation with industrial biotechnology and healthcare, advancing biofilm detection, sustainable food production, and medical diagnostics, bridging academia and industry for impactful, data-driven solutions.Professor Claudio Angione, Teesside University Professor Angione develops AI integrating molecular data and biological mechanisms, combining data, model, and knowledge-driven methods for health and biotechnology applications at the intersection of AI and biology.Mark Taylor, Managing Scientist at Marlow Foods, producers of Quorn Mark Taylor helps run a lab of cross functional disciplines, including genetics, analytics and fermentation. Marks particular interests include fermentation and scale up/down solutions.19'