Table 2 Performance comparison of different hypertension classification and diagnosis models across multiple facial Regions.

Model	Accuracy	Precision	Recall	F1	AUC
Face_ResNet−18	0.83	0.81	0.72	0.75	0.84
Face_ResNet−34	0.79	0.75	0.65	0.67	0.78
Face_ResNet−50	0.77	0.70	0.63	0.65	0.74
Forehead_ResNet−18	0.78	0.72	0.72	0.72	0.72
Forehead_ResNet−34	0.76	0.69	0.68	0.68	0.68
Forehead_ResNet−50	0.76	0.69	0.70	0.70	0.7
Zygoma_ResNet−18	0.80	0.76	0.68	0.70	0.75
Zygoma_ResNet−34	0.82	0.78	0.72	0.74	0.78
Zygoma_ResNet−50	0.79	0.74	0.66	0.68	0.73
Cheek_ResNet−18	0.82	0.79	0.70	0.73	0.76
Cheek_ResNet−34	0.79	0.79	0.62	0.64	0.68
Cheek_ResNet−50	0.80	0.90	0.62	0.63	0.69
Nose_ResNet−18	0.77	0.75	0.58	0.58	0.58
Nose_ResNet−34	0.75	0.67	0.56	0.55	0.6
Nose_ResNet−50	0.77	0.88	0.56	0.54	0.62
Lip_ResNet−18	0.73	0.61	0.58	0.58	0.75
Lip_ResNet−34	0.69	0.53	0.52	0.52	0.7
Lip_ResNet−50	0.69	0.56	0.55	0.55	0.68
Jaw_ResNet−18	0.77	0.88	0.56	0.54	0.57
Jaw_ResNet−34	0.77	0.71	0.61	0.62	0.64
Jaw_ResNet−50	0.75	0.88	0.52	0.47	0.58

Quick links

Search