Quantitative profiling of whole-brain projectomes at single-axon resolution using deep learning and high-resolution light sheet microscopy

MAPL3, an end-to-end deep learning pipeline for tera-voxel-scale light-sheet fluorescence microscopy, reconstructs brain-wide axon projections at single-axon resolution and generalizes across tracers, microscopes, and protocols.

• Why it matters: Understanding how individual axons build a brain-wide projectome is essential for linking circuit structure to function, but existing segmentation and tractography models struggle to generalize across imaging conditions.
• What they did: MAPL3 combines self-supervised learning with the SPECTRE network to segment axons and reconstruct projections from tera-voxel-scale light-sheet data, adding uncertainty estimates and a tractography module that quantifies passing versus terminating streamlines.
• The result: Applied to orbitofrontal cortex projectomes, MAPL3 reconstructs long-range streamlines that innervate established target regions and were missed by current pipelines, enabling subject- and population-level quantitative profiling.

AI summary of the abstract · may contain errors · not medical advice · read the paper.