Synapses display remarkable molecular and structural diversity that underlies neuronal connectivity and brain function. Super-resolution microscopy now enables imaging of their nanoscale organization within intact circuits, generating complex datasets that demand scalable analytical frameworks. Self-supervised and foundation models provide robust, annotation-free analysis, capturing latent representations of synaptic architecture for tasks such as segmentation and phenotypic clustering. Leveraging these models allows discovery of subtype-specific remodeling patterns and subtle nanoscale activity-dependent synaptic remodelling, advancing data-driven understanding of synaptic diversity.
Bio
Transdisciplinarity is at the heart of the research projects of the Flavie Lavoie-Cardinal team, combining optics and photonics, machine learning as well as cellular and molecular neuroscience. She is interested in the development of super-resolution microscopy and the use of machine learning methods to develop intelligent microscopes that can adapt to samples. These microscopes are then used for the study of molecular processes at the basis of communication and synaptic plasticity.