Metric scores used in <b>Multi-scale V-net architecture with deep feature CRF layers for brain extraction</b>

Park et al. investigate adding major improvements to a base segmentation model to increase robustness and go beyond the provided ground truth segmentation. Results indicate an overall improvement in the metrics and provide further discussion towards improving brain extraction models. This is the data that were used in creating the table and the plots.Park等人针对基础分割模型展开重大改进研究，旨在提升模型鲁棒性并突破现有标注真值（ground truth）分割的应用局限。实验结果表明，各项评估指标均实现整体提升，同时本研究还针对脑提取模型的优化方向展开了进一步探讨。本数据集即为用于生成该研究中表格与绘图的原始数据。