GBM multi-mode MR image segmentation method based on classifier weight converter

The invention discloses a GBM multi-modal MR image segmentation method based on a classifier weight converter, and the method comprises the steps: collecting a multi-modal nuclear magnetic resonance image, carrying out the slicing and standardization of the multi-modal nuclear magnetic resonance image in an upper computer, and obtaining an image matrix; fusing images of four modals in the image matrix into a tensor, and inputting the tensor into a brain tumor multi-modal MR image segmentation model to obtain a prediction probability distribution diagram; according to the method, a few-shot semantic segmentation method is used, the problems that a high-quality medical image data set is difficult to obtain and the data size is small can be effectively solved, the data collection and storage cost is further reduced to a certain extent, and therefore the GBM multi-mode MR image is efficiently segmented. 本发明公开了一种基于分类器权重转换器的GBM多模态MR图像分割方法，采集多模态核磁共振图像，然后在上位机中对多模态核磁共振图像经过切片处理与标准化处理获得图像矩阵，将图像矩阵中四个模态的图像融合成一个张量输入脑肿瘤多模态M