机器学习预测乳腺癌新辅助治疗后炎症代谢状态改变的模型评价
|
吴其蓁, 刘启明, 柴烨子, 陶政宇, 王依楠, 郭欣宁, 姜萌, 卜军
|
Evaluation of machine learning prediction of altered inflammatory metabolic state after neoadjuvant therapy for breast cancer
|
WU Qizhen, LIU Qiming, CHAI Yezi, TAO Zhengyu, WANG Yinan, GUO Xinning, JIANG Meng, PU Jun
|
|
表2 乳腺癌新辅助治疗前后患者5种模型在测试集中的单特征变量预测性能比较
|
Tab 2 Comparison of the predictive performance of the five models for single-feature variables in the test set for patients with breast cancer before and after neoadjuvant therapy
|
|
Single feature | Model | AUC | Accuracy | Precision rate | Recall | F1 score |
---|
WBC | RF | 0.715 | 0.603 | 0.656 | 0.636 | 0.646 | GB | 0.755 | 0.655 | 0.659 | 0.818 | 0.730 | SVM | 0.767 | 0.690 | 0.683 | 0.848 | 0.757 | DT | 0.655 | 0.569 | 0.643 | 0.545 | 0.590 | KNN | 0.787 | 0.690 | 0.703 | 0.788 | 0.743 | HB | RF | 0.835 | 0.845 | 0.853 | 0.879 | 0.866 | GB | 0.816 | 0.759 | 0.788 | 0.788 | 0.788 | SVM | 0.906 | 0.810 | 0.824 | 0.848 | 0.836 | DT | 0.748 | 0.759 | 0.788 | 0.788 | 0.788 | KNN | 0.846 | 0.793 | 0.862 | 0.758 | 0.806 | HDL | RF | 0.872 | 0.810 | 0.824 | 0.848 | 0.836 | GB | 0.853 | 0.828 | 0.829 | 0.879 | 0.853 | SVM | 0.889 | 0.845 | 0.833 | 0.909 | 0.870 | DT | 0.818 | 0.759 | 0.771 | 0.818 | 0.794 | KNN | 0.892 | 0.828 | 0.829 | 0.879 | 0.853 | IL-2R | RF | 0.709 | 0.672 | 0.684 | 0.788 | 0.732 | GB | 0.721 | 0.655 | 0.710 | 0.667 | 0.688 | SVM | 0.731 | 0.690 | 0.683 | 0.848 | 0.757 | DT | 0.627 | 0.638 | 0.676 | 0.697 | 0.687 | KNN | 0.719 | 0.655 | 0.651 | 0.848 | 0.737 | IL-8 | RF | 0.672 | 0.603 | 0.692 | 0.545 | 0.610 | GB | 0.661 | 0.690 | 0.727 | 0.727 | 0.727 | SVM | 0.330 | 0.569 | 0.569 | 1.000 | 0.725 | DT | 0.650 | 0.672 | 0.733 | 0.667 | 0.698 | KNN | 0.475 | 0.500 | 0.553 | 0.636 | 0.592 |
|
|
|