Convolutional kernels with an element-wise weighting mechanism for identifying abnormal brain connectivity patterns

作者：

Highlights：

• New CKEW can extract hierarchical topological features for brain networks.

• Our feature analysis method can find key features of brain networks as biomarkers.

• CKEW can obtain the highest classification accuracies among all testing methods.

摘要

•New CKEW can extract hierarchical topological features for brain networks.•Our feature analysis method can find key features of brain networks as biomarkers.•CKEW can obtain the highest classification accuracies among all testing methods.

论文关键词：Brain network classification,Convolutional kernels,Element-wise weighting mechanism,Topological features,Abnormal connectivity patterns

论文评审过程：Received 28 August 2019, Revised 26 June 2020, Accepted 28 July 2020, Available online 5 August 2020, Version of Record 15 August 2020.

论文官网地址：https://doi.org/10.1016/j.patcog.2020.107570