Estimating sparse functional connectivity networks via hyperparameter-free learning model

Highlights：

• Consequently, we propose a parameter-free method for FCN construction based on the global representation among fMRI time courses, which can automatically generate sparse FCNs.

• To verify the effectiveness of the proposed method, we conduct experiments to identify subjects with mild cognitive impairment and Autism spectrum disorder from normal controls based on the estimated FCNs.

• Experimental results on two benchmark databases demonstrate that the achieved classification performance of our proposed scheme is comparable to four conventional methods.

摘要

•Functional connectivity networks (FCNs) provide a potential way for understanding the brain organizational patterns and diagnosing neurological diseases. However, the parameter selection for estimating a sparse FCN is a challenging task.•Consequently, we propose a parameter-free method for FCN construction based on the global representation among fMRI time courses, which can automatically generate sparse FCNs.•To verify the effectiveness of the proposed method, we conduct experiments to identify subjects with mild cognitive impairment and Autism spectrum disorder from normal controls based on the estimated FCNs.•Experimental results on two benchmark databases demonstrate that the achieved classification performance of our proposed scheme is comparable to four conventional methods.

论文评审过程：Received 13 April 2020, Revised 14 October 2020, Accepted 15 December 2020, Available online 18 December 2020, Version of Record 29 December 2020.