Skip to main content

Table 3 Performance of individual DLMs for contrast-enhancement classification: comparison among the algorithm models in each body part

From: Development and validation of an ensemble artificial intelligence model for comprehensive imaging quality check to classify body parts and contrast enhancement

Body part Preprocessing Internal validation set External validation set
Precision (%) Recall (%) Accuracy (%) F1-score (%) Precision (%) Recall (%) Accuracy (%) F1-score (%)
Brain Mid-axial 100 100 100 100 100 100 100 100
Neck 100 100 100 100 98.38 98.33 98.33 98.3
Chest 100 100 100 100 98.38 98.48 98.41 98.4
Abdomen 100 100 100 100 100 100 100 100
Abdomen and pelvis 100 100 100 100 97.78 97.87 98.34 97.9
Brain Mid-sagittal 100 100 100 100 99.13 98.33 98.85 98.7
Neck 100 100 100 100 84.09 76.66 76.66 80
Chest 100 100 100 100 100 100 100 100
Abdomen 100 100 100 100 100 100 100 100
Abdomen and pelvis 100 100 100 100 100 100 100 100
Brain Mid-coronal 100 100 100 100 99.13 98.33 98.85 98.7
Neck 100 100 100 100 91.17 91.66 91.66 91.4
Chest 100 100 100 100 100 100 100 100
Abdomen 100 100 100 100 100 100 100 100
Abdomen and pelvis 100 100 100 100 100 100 100 100