Marking method and device and data processing method

The invention discloses a marking method and device and a data processing method. The method comprises the following steps: acquiring a first group of images, the first group of images comprising a plurality of images to be labeled; pre-segmenting the first group of images through a first segmentation model to obtain a pre-segmentation result; correcting a pre-segmentation result, and training the first segmentation model again based on the corrected pre-segmentation result to obtain a second segmentation model; and marking a second group of images to be marked through the second segmentation model to obtain a marking result of the second group of images. The technical problem that in the prior art, the aorta labeling speed in CT angiography is low is solved. 本发明公开了一种标注方法、装置以及数据处理方法。其中，该方法包括：获取第一组图像，其中，所述第一组图像包括待标注的多个图像；通过第一分割模型，对所述第一组图像进行预分割，得到预分割结果；对预分割结果进行修正，并基于修正后的所述预分割结果，对所述第一分割模型进行再次训练，得到第二分割模型；通过所述第二分割模型，对待标注的第二组图像进行标注，得到所述第二组图像的标注结果。本发明解决了现有技术中对CT血管造影中的主动脉标注速度较慢的技术问题。