基于多源数据融合的医用影像辅助诊断模型设计

Design of medical image⁃assisted diagnosis model based on multi⁃source data fusion

为了实现对患者的准确分类,辅助医生进行疾病诊断识别,文中提出一种基于多源数据融合的影像数据辅助诊断模型。该模型将MRI和PET图像进行融合,并以改进的Transformer网络T2T-ViT为主干分类网络,通过迁移学习ImageNet数据集的参数,实现对阿尔茨海默病的分类。在公开数据集上进行的实验结果表明,所提出模型对于阿尔茨海默症患者的识别准确率可达0.95,优于目前的主流图像分类网络,证明其有效性,能够辅助影像医生进行疾病诊断,具有一定的应用价值。

An image data assisted diagnosis model based on multi⁃source data fusion is proposed to achieve accurate classification of patients and assist doctors in disease diagnosis and recognition.In this model,MRI(magnetic resonance imaging)and PET(positron emission tomography)images are fused,and the improved transformer network T2T⁃ViT is taken as the backbone classification network.By transfer learning the parameters of ImageNet data set,the Alzheimer's disease(AD)is classified.The experimental results on the public dataset show that the recognition accuracy of the proposed model for patients with AD can reach 0.95,which is higher than that of the current mainstream image classification network.This proves that the proposed model is effective,and it can assist imaging doctors in disease diagnosis.Therefore,it has a certain application value.