Recipient organisationNIHR Birmingham Biomedical Research Centre
NIHR supportRecorded as supported by this research centre
PeriodJan 2025 — Jul 2026
In plain English
AI plain-English summary
A fully automated microscope platform will use artificial intelligence to capture and analyse living cells in five dimensions—3D space, time, and multiple fluorescent labels—without human intervention. Current advanced microscopy generates too little data to train AI models effectively, because the process is slow and requires a skilled operator at every step. This bottleneck blocks the use of machine learning to understand how cells signal to each other. The researchers will build a platform that autonomously acquires and analyses huge 5D datasets at high resolution, creating the training data that AI needs. If successful, the platform will allow scientists to study G protein-coupled receptors—the largest family of cell receptors and a major target for existing drugs—in unprecedented detail. This could reveal how these receptors work in heart failure, cancer, and diabetes, and open new routes for drug development. The work is fundamental science: it aims to answer basic questions about cell signalling rather than deliver a therapy directly. But similar fundamental advances in microscopy and AI have repeatedly led to unexpected clinical breakthroughs.
View original technical description
Fluorescence microscopy has emerged as one of the most powerful approaches to investigate the fundamental mechanisms of life, including the complex dynamic events allowing our cells to respond to external and internal stimuli like hormones and neurotransmitters, by using fluorescent labels to tag molecules of interest and follow them over time. The application of machine learning (ML) and artificial intelligence (AI) to advanced fluorescence microscopy presents immense potential to model and understand the complex mechanisms of cellular signalling with unparalleled depth, paving the way to transformative scientific discoveries and innovative therapies for conditions like heart failure, cancer or diabetes. However, the widespread application of ML/AI to discovery science is severely hampered by a lack of training data resulting from the low-throughput and heavy user-dependence of current advanced microscopy approaches. To overcome these limitations, we will develop an innovative, AI-driven, fully automated high-throughput platform, seamlessly integrating high-throughput 5D fluorescence imaging with comprehensive automation, deep learning-based data analysis, and generative modelling of complex cellular structures and biological processes. The platform will be capable of autonomously acquiring and analysing very large datasets of living cells in 5D (3D space, time, and multiple fluorescent labels) at high spatiotemporal resolution. Leveraging this platform, we will systematically investigate how G protein-coupled receptors (GPCRs), the largest family of cell receptors deeply involved in human physiology and disease, work with unprecedented detail, addressing fundamental and still unanswered scientific questions and unlocking novel avenues for drug development.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know