Kirilina, E. | bioRxiv Science

bioRxiv2018

Apparent thinning of visual cortex during childhood is associated with myelination, not pruning

Microstructural mechanisms underlying apparent cortical thinning during childhood development are unknown. Using functional, quantitative, and diffusion magnetic resonance imaging in children and adults, we tested if tissue growth (lower T1 relaxation time and mean diffusivity (MD)) or pruning (higher T1 and MD) underlies cortical thinning in ventral temporal cortex (VTC). After age 5, T1 and MD decreased in mid and deep cortex of functionally-defined regions in lateral VTC, and in their adjacent white matter. T1 and MD decreases were (i) consistent with tissue growth related to myelin proliferation, which we verified with adult postmortem histology and (ii) correlated with apparent cortical thinning. Thus, contrary to prevailing theories, cortical tissue does not thin during childhood, it becomes more myelinated, shifting the gray-white matter boundary deeper into cortex. As tissue growth is prominent in regions with protracted functional development, our data suggest an intriguing hypothesis that functional development and myelination are interlinked.

neuroscience↗

bioRxiv2018

Deep Learning meets Topology-preserving Active Contours: towards scalable quantitative histology of cortical cytoarchitecture.

Deep learning has thoroughly changed the field of image analysis yielding impressive results whenever enough annotated data can be gathered. While partial annotation can be very fast, manual segmentation of 3D biological structures is tedious and error-prone. Additionally, high-level shape concepts such as topology or boundary smoothness are hard if not impossible to encode in Feedforward Neural Networks. Here we present a modular strategy for the accurate segmentation of neural cell bodies from light-sheet microscopy combining mixed-scale convolutional neural networks and topology-preserving geometric deformable models. We show that the network can be trained efficiently from simple cell centroid annotations, and that the final segmentation provides accurate cell detection and smooth segmentations that do not introduce further cell splitting or merging.

bioinformatics↗