ADHD classification by dual subspace learning using resting-state functional connectivity
作者：
Highlights：
• We use a subspace learning model to form two subspaces to separate ADHD components from healthy control ones.
• Several subspace measures are employed as kernels to enhance the component clustering performance.
• A novel ADHD classification framework is given via using a binary hypothesis testing of test data.
• The features of training data are generated by using the functional connectivity of test data with its label hypothesis.
• The projected feature energies of training data are compared under binary hypotheses to predict ADHD subjects.
摘要
•We use a subspace learning model to form two subspaces to separate ADHD components from healthy control ones.•Several subspace measures are employed as kernels to enhance the component clustering performance.•A novel ADHD classification framework is given via using a binary hypothesis testing of test data.•The features of training data are generated by using the functional connectivity of test data with its label hypothesis.•The projected feature energies of training data are compared under binary hypotheses to predict ADHD subjects.
论文关键词：ADHD,Feature selection,Graph embedding,Graph Laplacian,Subspace learning,SVM-RFE
论文评审过程：Received 10 April 2019, Revised 11 December 2019, Accepted 30 December 2019, Available online 13 January 2020, Version of Record 23 January 2020.
论文官网地址：https://doi.org/10.1016/j.artmed.2019.101786
