Multi-modal neural image tumor segmentation method and system based on deep learning

The invention discloses a multi-modal neural image tumor segmentation method based on deep learning, and the method comprises the steps: obtaining a PET-CT image pair composed of a PET image and a corresponding CT image, carrying out the preprocessing of the PET-CT image pair, and obtaining a preprocessed PET-CT image pair, and inputting the preprocessed PET-CT image pair into the trained multi-modal neural image tumor segmentation model to obtain a final focus segmentation prediction result. According to the method, the technical problem that the segmentation result is inaccurate due to the fact that the weight and feature differences between different modals are ignored in the existing method for performing segmentation after bitwise addition and fusion, information of some modals is excessively emphasized, and information of other modals is ignored can be solved. 本发明公开了一种基于深度学习的多模态神经图像肿瘤分割方法，包括：获取正电子发射型计算机断层显像PET图像及其对应的计算机断层扫描CT图像所组成的PET-CT图像对，对该PET-CT图像对进行预处理，以得到预处理后的PET-CT图像对，将预处理后的PET-CT图像对输入到训练好的多模态神经图