Table 5 Performance evaluation of selected models.

From: Explainable artificial intelligence for stroke prediction through comparison of deep learning and machine learning models

Models

Sensitivity (%)

Specificity (%)

Accuracy (%)

F1-score (%)

ROC (%)

Performance of each deep learning models

CNN

96.0

87.0

92.0

91.0

96.0

LSTM

96.15

80.0

89.47

87.34

97.0

DNN

93.0

93.0

95.0

93.0

95.0

FNN

94.0

96.0

96.0

95.0

98.0

Performance of each machine learning models

RF

99.9

100

99.0

99.0

99.0

KNN

70.0

94.0

86.0

80.0

83.0

XGB

94.0

95.0

95.0

95.0

95.0

SVM

29.0

94.0

71.0

44.0

64.0