Queen Mary University of London scientists develop AI-powered software for efficient cell tracking

Deep learning, a branch of machine learning and artificial intelligence, is transforming cell dynamics research, providing scientists with powerful new tools to study the complex movements of cells and their components. In a comprehensive review article published in Trends in Cell Biology, Professor Viji M. Draviam Sastry and Dr. Binghao Chai from Queen Mary University of London explore the opportunities and challenges of using deep learning to analyse microscopy data, highlighting the potential of this technology to apply to the field.?

"Deep learning is rapidly changing the way we study cell dynamics," said Viji Draviam, Queen Mary’s Professor of Quantitative Cell and Molecular Biology and Director of Industrial Innovation. "By enabling us to automatically track and analyse the movements of cells and their components in high-throughput microscopy experiments, deep learning is providing us with new insights into cellular processes that were previously inaccessible."?

The authors of the review article developed SpinX, an open-source deep learning-based software tool that can automatically identify and track moving objects in 3D microscopy data. SpinX is available through the Zeiss/arivis cloud platform, making it easy for researchers to implement and use.?

"SpinX is a powerful tool that can save researchers a significant amount of time and effort," said Dr. Chai. "It can automatically track objects in complex microscopy data, allowing researchers to focus on analysing the results and making new discoveries."?

In addition to discussing the potential of deep learning for cell dynamics research, the review article also identifies some of the challenges that remain in the field. One challenge is the need to develop more robust and reliable deep learning algorithms that can work with a wider range of microscopy data – for example, discontinuous time-lapse movies that capture cell dynamics. Another challenge is the need to make deep learning tools more accessible to researchers who are not experts in machine learning.?

"Despite these challenges, we are confident that deep learning will continue to play an increasingly important role in cell dynamics research," said Professor Draviam. "We are committed to developing new deep learning tools that will help researchers make groundbreaking discoveries through drug development and genotype-phenotype matching screens which consume enormous time and effort."?