Background Vasogenic cerebral edema resulting from blood–brain barrier (BBB) damage aggravates the devastating consequences of intracerebral hemorrhage (ICH). Although augmentation of endothelial Wnt/β‐catenin signaling substantially alleviates BBB breakdown in animals, no agents based on this mechanism are clinically available. Lithium is a medication used to treat bipolar mood disorders and can upregulate Wnt/β‐catenin signaling. Methods We evaluated the protective effect of lithium on the BBB in a mouse model of collagenase IV‐induced ICH. Furthermore, we assessed the effect and dependency of lithium on Wnt/β‐catenin signaling in mice with endothelial deletion of the Wnt7 coactivator Gpr124. Results Lithium treatment (3 mmol/kg) significantly decreased the hematoma volume (11.15 ± 3.89 mm3 vs. 19.97 ± 3.20 mm3 in vehicle controls, p = 0.0016) and improved the neurological outcomes of mice following ICH. Importantly, lithium significantly increased the BBB integrity, as evidenced by reductions in the levels of brain edema (p = 0.0312), Evans blue leakage (p = 0.0261), and blood IgG extravasation (p = 0.0009) into brain tissue around the hematoma. Mechanistically, lithium upregulated the activity of endothelial Wnt/β‐catenin signaling in mice and increased the levels of tight junction proteins (occludin, claudin‐5 and ZO‐1). Furthermore, the protective effect of lithium on cerebral damage and BBB integrity was abolished in endothelial Gpr124 knockout mice, suggesting that its protective effect on BBB function was mainly dependent on Gpr124‐mediated endothelial Wnt/β‐catenin signaling. Conclusion Our findings indicate that lithium may serve as a therapeutic candidate for treating BBB breakdown and brain edema following ICH.

Semantic segmentation has always been a fundamental and critical task to scene understanding. Current deep convolutional neural networks (DCNN) are able to successfully learn context from very large receptive fields due to convolutions with deep layers. However, current convolutions in DCNNs does not consider local object boundaries that are the borders among different semantic regions. Convolution with equal contribution on the pixels across the boundary may lead to inferior segmentation results. In this paper, a novel boundary-aware convolution is proposed. It is able to effectively fuse features by adaptively assigning contributions from pixels within receptive fields according to the boundary similarity map. A new semantic segmentation network based on classical FCN8S is then designed by employing multi-scale boundary-aware convolution. The whole network is implemented end-to-end and evaluated with heterogeneous RGB and depth input. Experiments conducted on multiple datasets show that our boundary-aware CNN can effectively improve the semantic segmentation performance.

No abstract