A New Approach for Clustered MCs Classification with Sparse Features Learning and TWSVM

Table 1

Experimental results of the proposed MCs-SRC method for MCs detection, compared with sparse representation based TWSVMs and SVMs methods.

Methods	Sensitivity	Specificity	Az
MCs-SRC	90.84±1.07%	92.37±0.78%	0.9407±0.0564
TWSVMs-SR	92.07±0.89%	89.93±0.91%	0.9678±0.0977
SVMs-SR	87.53±0.94%	89.72±0.88%	0.9304±0.1001