SUMMARY Experience rearranges anatomical connectivity in the brain, but such plasticity is suppressed in adulthood. We examined the turnover of dendritic spines and axonal varicosities in the somatosensory cortex of mice lacking Nogo Receptor 1 (NgR1). Through adolescence, the anatomy and plasticity of ngr1 null mice are indistinguishable from control, but suppression of turnover after age 26 days fails to occur in ngr1−/− mice. Adolescent anatomical plasticity can be restored to one-year old mice by conditional deletion of ngr1. Suppression of anatomical dynamics by NgR1 is cell autonomous, and is phenocopied by deletion of Nogo-A ligand. Whisker removal deprives the somatosensory cortex of experience-dependent input and reduces dendritic spine turnover in adult ngr1−/− mice to control levels, while an acutely enriched environment increases dendritic spine dynamics in control mice to the level of ngr1−/− mice in a standard environment. Thus, NgR1 determines the low set point for synaptic turnover in adult cerebral cortex.

No abstract

In tasks involving the interpretation of medical images, suitably trained machine-learning models often exceed the performance of medical experts. Yet such a high-level of performance typically requires that the models be trained with relevant datasets that have been painstakingly annotated by experts. Here we show that a self-supervised model trained on chest X-ray images that lack explicit annotations performs pathology-classification tasks with accuracies comparable to those of radiologists. On an external validation dataset of chest X-rays, the self-supervised model outperformed a fully supervised model in the detection of three pathologies (out of eight), and the performance generalized to pathologies that were not explicitly annotated for model training, to multiple image-interpretation tasks and to datasets from multiple institutions.