Skip to main content
. 2026 Sep 22;9:1934912. doi: 10.3389/frai.2026.1934912

Table 13.

Ablation analysis of lag-only, lead-only, and proposed Lag-Lead compensation.

Model name Precision Recall F1-score Accuracy Confusion matrix
EfficientNetB3 0.88 0.87 0.81 0.90 Confusion matrix graphic displaying true labels versus predicted labels for benign and malignant cases. Top left shows 114 true negatives, top right has 10 false positives, bottom left has 6 false negatives, and bottom right has 34 true positives.
Lag compensation 0.82 0.85 0.84 0.92 Confusion matrix for a classification model with true labels on the y-axis and predicted labels on the x-axis, showing 117 true benign, 7 false positive, 6 false negative, and 34 true malignant classifications.
Lead compensation 0.91 0.77 0.83 0.92 Confusion matrix with two classes, benign and malignant, showing 121 true benign predictions, 3 benign cases classified as malignant, 9 malignant cases classified as benign, and 31 true malignant predictions.
Proposed model without spatial filters 0.90 0.67 0.77 0.90 Confusion matrix graphic for a classification model, showing counts for actual versus predicted categories: 121 true benign, 3 benign predicted as malignant, 13 malignant predicted as benign, 27 true malignant.
Proposed model 0.89 0.90 0.81 0.90 Confusion matrix showing model classification results with actual labels on the y-axis and predicted labels on the x-axis; 112 true benign, 12 false positive, 4 false negative, and 36 true malignant cases.

Bold values represent the metric values obtained by the proposed model by combining all the components.