Enabling Live-Cell 4D Super-Resolution Microscopy Guided by Artificial Intelligence

ongoing

Most fundamental biomedical discoveries are carried by indirect observation, due to our inability to follow molecular-level cellular processes non-invasively in living samples. There is a major need to develop the capacity to directly observe the molecular basis of cell regulation at relevant spatial and temporal scales. My work has focused on establishing Next-Generation Super-Resolution Microscopy able to accurately describe the nanoscale structure of living-cells, beyond the capacity of conventional imaging. Despite significant advancements, Super-Resolution Microscopy does not allow observations for more than a few minutes before damaging high-intensity illumination compromises cell behaviour. Better fluorophores and optics can partly address this problem; however, a fundamental limitation remains - humans drive imaging. In microscopy, researchers take educated guesses on how to observe a sample based on empirical criteria. Acquisition settings are then kept static, despite the diversity of spatiotemporal scales associated with cell behaviour in health and disease. This project will solve these challenges by establishing self-driving microscopes, able to adapt in real-time to the biological phenomenon under observation. Doing so, I will enable the capacity for unprecedented 4D imaging data optimised for content, resolution and quality while remaining non-invasive for long periods of time. To this end, I will bridge and evolve cutting-edge concepts in Computational Optical Microscopy and Machine Learning, effectively establishing Machine Learning Guided 4D Super-Resolution Microscopy. These approaches will enable 4D live-cell nanoscopic imaging over record periods and challenge the assumption that microscopy needs to obey homogeneous temporal sampling. Its enabling capacity will be demonstrated by visualising nanoscale cellular events previously unseen over hours, such as the molecular-level progression of viral infection.