DIAS
收藏资源简介:
DIAS数据集是由北京邮电大学人工智能学院创建,专门用于DSA序列中的颅内动脉分割。该数据集包含120个序列,共计753帧高分辨率(800×800像素)图像,涵盖了从4到14帧的不同长度序列。数据集的创建过程涉及由神经外科医生对像素级别进行细致标注,确保了数据的高质量和准确性。DIAS数据集的应用领域包括计算机辅助的卒中研究和临床实践,特别是在量化血管形态学方面发挥关键作用。
The DIAS dataset was developed by the School of Artificial Intelligence, Beijing University of Posts and Telecommunications, specifically for intracranial artery segmentation in DSA sequences. It includes 120 sequences, totaling 753 high-resolution (800×800 pixels) image frames, with sequence lengths ranging from 4 to 14 frames. The dataset creation involved meticulous pixel-level annotations performed by neurosurgeons, ensuring the high quality and accuracy of the data. Applications of the DIAS dataset cover computer-aided stroke research and clinical practice, where it plays a critical role particularly in quantifying vascular morphology.




