基于自注意力机制的阿尔茨海默病预测研究

A PREDICTIVE MODEL FOR ALZHEIMER'S DISEASE BASED ON SELF-ATTENTION MECHANISM

为了更加准确地预测阿尔茨海默病(Alzheimer’s Disease,AD),提出一种基于自注意力机制的预测模型。对核磁共振图像(Magnetic Resonance Imaging,MRI)进行预处理并提取人脑解剖结构的初级特征,设计基于自注意力机制的特征处理单元并基于残差结构设计模式构建可靠的网络架构,自动解析人脑解剖结构的依赖关系并生成MRI图像的有效特征表示,进而实现对AD的预测。实验结果表明,该模型对AD的分类准确率为99.36%,对轻度认知障碍(Mild Cognitive Impairment,MCI)的分类准确率为98.90%。与现有方法比较,该模型拥有更好的预测性能。

A self-attention mechanism based model is proposed for the prediction of Alzheimer s disease(AD).Magnetic resonance imaging(MRI)images were pre-processed to extract primary features for brain anatomical structures.A self-attention mechanism based feature processing unit(SAFPU)was designed,and by the theory of residual blocks,multiple SAFPUs were stacked to build a reliable network for predicting AD,which could automatically analyze the dependencies of different brain anatomical structures to generate high-level features for MRI images.The empirical results demonstrate the proposed model outperforms existing AD classification methods,which achieves 99.36%(98.90%)of the maximum accuracy for the AD(early stage of AD,i.e.,Mild Cognitive Impairment)classification task.