The image contains two bar graphs. The first graph, labeled (a), shows the performance of a proposed Physics-Guided Unified Damage Index framework using repeated stratified five-fold cross-validation. The x-axis represents evaluation metrics: Accuracy, Precision, Recall, and F1-score. The y-axis represents performance in percentage. The bars indicate mean performance values with error bars showing standard deviations. Accuracy is approximately 88.85 percentage with a standard deviation of 8.04 percentage, Precision is approximately 91.37 percentage with a standard deviation of 6.32 percentage, Recall is approximately 93.58 percentage with a standard deviation of 8.68 percentage, and F1-score is approximately 92.18 percentage with a standard deviation of 6.00 percentage. The second graph, labeled (b), displays bootstrap-based 95 percentage confidence intervals for the performance metrics. The x-axis again represents evaluation metrics: Accuracy, Precision, Recall, and F1-score.(a) Performance of the proposed Physics-Guided Unified Damage Index framework evaluated using repeated stratified five-fold cross-validation. The model achieved mean accuracy, precision, recall, and F1-score of 88.85 ± 8.04%, 91.37 ± 6.32%, 93.58 ± 8.68%, and 92.18 ± 6.00%, respectively, demonstrating stable predictive performance and good generalization across multiple data partitions. (b) Bootstrap-based 95% confidence intervals for the performance metrics of the proposed anomaly detection framework. The estimated confidence intervals indicate consistent predictive performance while quantifying the uncertainty associated with the relatively small testing dataset, providing additional statistical evidence supporting the robustness of the proposed methodology