Research Article

An Efficient Retinal Segmentation-Based Deep Learning Framework for Disease Prediction

Table 4

Comparison results of existing with proposed method.

Existing methodsAlom [4]Wang [18]Jiang [11]Guo [34]Proposed
MethodsRU-NetR2U-NetDense U-netFCNBSCNEDLFDPRS
DatasetDSCDSCDSDSCDSCDSC

ACC95.5697.0696.2295.5697.1296.3495.1195.3896.4696.3397.798.4698.7298.8997.3297.3397.44
score91.5593.9678.191.7184.7592.8——76.7577.558282.3685.4775.683.8592.381.94
Sensitivity77.5181.0874.5977.9282.9877.5679.8679.1466.6383.3283.2381.7987.5179.7282.5682.1283.92
Specificity98.1698.7198.3698.1398.6298.297.3697.2299.3397.498.6798.7998.9498.9698.6898.5798.45
AUC97.8299.0998.0397.8499.1498.1597.497.0497.897.999.1298.7499.4198.7498.7998.8498.79

Note: D: DRIVE; S: STARE; C: CHASE; ACC: accuracy; sensitivity; specificity; AUC: area under the ROC curve.