Subject-adaptive Integration of Multiple SICE Brain Networks with Different Sparsity
作者：
Highlights：
• In this paper, we propose a learning based framework that integrates a set of SICE networks with the aim of attaining more discriminative power. Our framework has at least four contributions.
• It makes use of the whole spectrum of SICEs to improve the accuracy and circumvent presetting the employed sparsity level.
• The proposed framework provides subject-adaptive integration of SICE networks.
• Our integration of SICE networks respects the specific geometric property of SICE matrix.
• By the integration, our learning framework provides a unique, enhanced, and new network representation for each subject.
摘要
HighlightsIn this paper, we propose a learning based framework that integrates a set of SICE networks with the aim of attaining more discriminative power. Our framework has at least four contributions.•It makes use of the whole spectrum of SICEs to improve the accuracy and circumvent presetting the employed sparsity level.•The proposed framework provides subject-adaptive integration of SICE networks.•Our integration of SICE networks respects the specific geometric property of SICE matrix.•By the integration, our learning framework provides a unique, enhanced, and new network representation for each subject.
论文关键词：Brain network integration,Sparse representation classification,Subject-adaptive learning,Brain disease diagnosis
论文评审过程：Received 1 February 2016, Revised 7 September 2016, Accepted 21 September 2016, Available online 22 September 2016, Version of Record 27 November 2016.
论文官网地址：https://doi.org/10.1016/j.patcog.2016.09.024
