Skip to main content
. 2026 Sep 28;16:1849274. doi: 10.3389/fonc.2026.1849274

Figure 1.

Figure consists of three ROC curve graphs (A, B, C) and three normalized confusion matrices (D, E, F) comparing a model’s performance across training, internal, and external test sets. ROC curves show AUCs of 0.836, 0.80, and 0.704 respectively, indicating decreasing predictive performance. Confusion matrices display proportions of true and predicted alive and dead cases; training and internal sets show higher accuracy, while external set performance drops, with lower values for correct alive and dead predictions.

Development and validation of the prognostic prediction model based on the CLAM self-attention framework. The model was validated using 10 independent iterations of stratified k-fold cross-validation. Normalized confusion matrices were used to assess model performance across different datasets. (A) ROC curve of the optimal model in the TCGA training set. (B) Performance of the optimal model in the TCGA internal test set. (C) Performance of the model in the external test set from the proprietary dataset of Ansteel General Hospital. (D) Normalized confusion matrix of the optimal model in the training set. (E) Normalized confusion matrix of the internal test set. (F) Normalized confusion matrix of the external test set.