. Author manuscript; available in PMC: 2019 May 15.

Published in final edited form as: Med Image Comput Comput Assist Interv. 2018 Sep 26;11070:647–654. doi: 10.1007/978-3-030-00928-1_73

Table 1.

Comparison with state-of-the-art methods for ASD identification using rs-fMRI ABIDE data. FNC: Functional Network Connectivity; KNN: k-nearest neighbor algorithm.

Methods	Feature type	Classifier	ACC (%)	SEN (%)	SPE (%)	AUC (%)
sGCN-1 [3]	FNC	KNN	60.50	–	–	57.00
sGCN-2 [3]	FNC	KNN	63.50	–	–	61.00
FCA [4]	FNC	t-test	63.00	72.00	58.00	–
DAE [5]	FNC	Softmax regression	66.00	66.00	65.00	–
MCLRR (ours)	FNC	KNN	69.10	70.24	66.43	68.33