Deep learning-driven multi-view multi-task image quality assessment method for chest CT image

Table 5 The multiple evaluation metrics results of the three control groups on three datasets

Dataset		Precision	Sensitivity	F1-score	P-value
Additional test set	M\(^2\)IQA VS GT	0.91	0.81	0.85	0.61
	MOS VS GT	0.88	0.85	0.86	0.42
	M\(^2\)IQA VS MOS	0.94	0.86	0.90	0.27
LungCT-Diagnosis	M\(^2\)IQA VS GT	1.00	0.83	0.91	0.73
	MOS VS GT	1.00	0.67	0.80	0.54
	M\(^2\)IQA VS MOS	0.70	0.88	0.78	0.85
CMB-LCA	M\(^2\)IQA VS GT	0.86	0.75	0.80	0.45
	MOS VS GT	0.86	0.75	0.80	0.36
	M\(^2\)IQA VS MOS	0.86	0.86	0.86	0.90

ISSN: 1475-925X