Table 3.

The comparison of achieved results with state-of-the-arts.

Task Methods Subjects number Performance
Prasad [37] Haar wavelet 89 Accuracy:97.8% Sensitivity:97.5% Specificity:97.75%
Omar [38] Multiscale LBP texture 130 Accuracy:98.68% Sensitivity:94.81% Specificity: 96.73%
Jian Wang [13] 2D MF-DFA-LSSVM 130 Accuracy:99.01% Sensitivity:99.03% Specificity: 97.73%
Koh [39] Continuous wavelet 910 Accuracy:92.48% Sensitivity:89.37% Specificity:95.58%
Acharya [40] Features from Gabor 510 Accuracy:93.10% Sensitivity:89.75% Specificity:96.20%
Stevenson [41] Artificial intelligence algorithms 4435 Accuracy:89.00% Sensitivity:75.00% Specificity:89.00%
Maheshwari [42] VMD and entropy 488 Accuracy:95.19% Sensitivity:93.62% Specificity:96.71%
Giraddi [43] Haar wavelet and First order statistical features 130 Accuracy:85% Sensitivity:87% Specificity:80%
Ours 2D EMD and Multifractal 60 Accuracy:96.24% Sensitivity:95.56%
Specificity:95.38%