ZHONGLIU YINGXIANGXUE

The application value of radiomics-based prediction of Bcl-2 and c-Myc expression status in patients with intracranial primary central nervous system lymphoma

LI Xie, HUANG Haitao, LI Huihu, LI Fan, WAN Yun

ZHONGLIU YINGXIANGXUE, 2025, 34(3): 273-281. DOI: 10.19732/j.cnki.2096-6210.2025.03.011

项目	AUC	准确度	精确度	召回度	F1分数	灵敏度	特异度	DeLong P值
XGBoost
训练集	0.96	0.93	0.92	0.94	1.00	0.90	1.00	0.00
内部验证集	0.90	0.91	0.92	0.86	0.89	0.86	0.94	0.02
外部验证集	0.88	0.88	0.81	0.94	0.87	0.94	0.82	0.17
GBM
训练集	0.88	0.85	0.73	0.96	0.84	0.98	0.77
内部验证集	0.83	0.81	0.72	0.92	0.81	0.93	0.72
外部验证集	0.91	0.90	0.82	0.94	0.90	1.00	0.82

Tab. 3 Predictive Performance of the Model

Other figure/table from this article

Tab. 1 Clinical and imaging data of the training set, internal validation set, and external validation set n（%）或$\bar{x}±s$
Fig.1 Flowchart for inclusion of participants
Tab. 2 Selected radiomic features
Fig.2 Analysis of correlation and importance of radiomic features
Fig.3 Ranking of multiple models by average AUC of internal and external validation sets