基于图的特征选择算法在阿兹海默症诊断问题研究

Graph feature selection for Alzheimer's disease diagnosis

针对神经性疾病难以确诊的问题,提出了一种基于图的特征选择方法,过滤掉不相干的特征,从而方便并且准确地对疾病患者进行诊断。算法首先基于先验知识定义了两种基本关系(特征关系和样本关系);然后将这两种关系嵌入到一个由最小二次损失函数和l2-范数正则化因子组成的多任务学习框架中进行特征选择;最后,将约简得到的降维矩阵送入支持向量机(SVM)中对阿兹海默症患者进行确诊。通过对Alzheimer’s disease neuroimaging initiative(ANDI)的研究数据集进行实验得知,提出算法的分类效果均优于一般常用分类算法,如K最近邻法(KNN)、支持向量机(SVM)等。提出的算法通过考虑特征选择和引入两种数据的内在关系,有效提高了阿兹海默疾病诊断的正确率。

This paper proposed a graph feature selection method for Alzheimer＇ s disease diagnosis, by adding the information inherent in the observations into a sparse multi-task learning framework. Specifically, this paper firstly defined two relations （i. e. , the feature-feature relation and the sample-sample relation, respectively） based on the prior knowledge. Then embed- ding these two kinds of relations into a multi-task learning framework （ i. e. , a least square loss function plus an 12-norm regularization term） to conduct feature selection. Furthermore, it fed the reduced data into a support vector machine （SVM） for conducting the identification of Alzheimer＇ s disease （AD）. Finally, the experimental results on a subset of the Alzheimer＇ s disease neuroimaging initiative （ADNI） dataset show the effectiveness of the proposed method in terms of classification accura- cy, by comparing with the state-of-the-art methods, including K nearest neighbor （KNN）, SVM, and so on. This paper proves that the proposed method can improve the performance of AD diagnosis accuracy, by take feature selection and two kinds of relation into account simultaneously.