Applying probability calibration to ensemble methods to predict 2-year mortality in patients with DLBCL.
Applying probability calibration to ensemble methods to predict 2-year mortality in patients with DLBCL.
复制标题
DOI：
10.1186/s12911-020-01354-0
复制
发表时间：
2021-01-07
影响因子：
3.5
通讯作者：
Luo Y
中科院分区：
文献类型：
--
作者：
Fan S;Zhao Z;Yu H;Wang L;Zheng C;Huang X;Yang Z;Xing M;Lu Q;Luo Y
在化疗方案、临床分期、免疫表达等因素的影响下，弥漫性大b细胞淋巴瘤（DLBCL）患者的生存率存在差异。准确预测死亡风险是精准医疗的关键，它可以帮助临床医生做出最佳的治疗决策，延长DLBCL患者的生存时间。因此，我们建立了一个预测模型来预测DLBCL患者治疗2年内的死亡风险。我们评估了406例DLBCL患者，并从每位患者收集了17个变量。预测变量选择采用Cox模型、logistic模型和随机森林算法。选择了五个分类器作为集成学习的基础模型：naïve贝叶斯、逻辑回归、随机森林、支持向量机和前馈神经网络模型。我们首先使用概率校准方法（包括形状限制多项式回归、Platt标度和等压回归）校准五个基本模型的偏置输出。然后，我们汇总各种基础模型的输出，采用三种策略（叠加、简单平均和加权平均）预测DLBCL患者的2年死亡率。最后，我们在300多个测试中评估了模型的性能。性别、分期、IPI、KPS和利妥昔单抗是预测DLBCL患者治疗2年内死亡的重要因素。采用形状限制多项式回归对基础模型进行首次校正的叠加模型在所有方法中表现最佳（AUC = 0.820, ECE = 8.983, MCE = 21.265）。相比之下，未进行概率校准的叠加模型性能较差（AUC = 0.806, ECE = 9.866, MCE = 24.850）。在简单平均模型和加权平均模型中，集合模型的预测误差也随着概率校正而减小。在所有比较的方法中，所提出的模型在预测DLBCL患者2年死亡率时预测误差最小。这些有希望的结果可能表明我们将概率校准应用于集成学习的建模策略是成功的。
Under the influences of chemotherapy regimens, clinical staging, immunologic expressions and other factors, the survival rates of patients with diffuse large B-cell lymphoma (DLBCL) are different. The accurate prediction of mortality hazards is key to precision medicine, which can help clinicians make optimal therapeutic decisions to extend the survival times of individual patients with DLBCL. Thus, we have developed a predictive model to predict the mortality hazard of DLBCL patients within 2 years of treatment. We evaluated 406 patients with DLBCL and collected 17 variables from each patient. The predictive variables were selected by the Cox model, the logistic model and the random forest algorithm. Five classifiers were chosen as the base models for ensemble learning: the naïve Bayes, logistic regression, random forest, support vector machine and feedforward neural network models. We first calibrated the biased outputs from the five base models by using probability calibration methods (including shape-restricted polynomial regression, Platt scaling and isotonic regression). Then, we aggregated the outputs from the various base models to predict the 2-year mortality of DLBCL patients by using three strategies (stacking, simple averaging and weighted averaging). Finally, we assessed model performance over 300 hold-out tests. Gender, stage, IPI, KPS and rituximab were significant factors for predicting the deaths of DLBCL patients within 2 years of treatment. The stacking model that first calibrated the base model by shape-restricted polynomial regression performed best (AUC = 0.820, ECE = 8.983, MCE = 21.265) in all methods. In contrast, the performance of the stacking model without undergoing probability calibration is inferior (AUC = 0.806, ECE = 9.866, MCE = 24.850). In the simple averaging model and weighted averaging model, the prediction error of the ensemble model also decreased with probability calibration. Among all the methods compared, the proposed model has the lowest prediction error when predicting the 2-year mortality of DLBCL patients. These promising results may indicate that our modeling strategy of applying probability calibration to ensemble learning is successful.
登录
查看更多内容
影响因子：
45.3
作者：
Fu, Kai;Weisenburger, Dennis D.;Vose, Julie M.
通讯作者：
Vose, Julie M.
影响因子：
51.1
作者：
Pfreundschuh, Michael;Trumper, Lorenz;Loeffler, Markus
通讯作者：
Loeffler, Markus
Ki-67 是淋巴瘤的一个有价值的预后预测因子，但其效用因淋巴瘤亚型而异：来自系统荟萃分析的证据
影响因子：
3.8
作者：
He X;Chen Z;Fu T;Jin X;Yu T;Liang Y;Zhao X;Huang L
通讯作者：
Huang L
影响因子：
--
作者：
AYER, M;BRUNK, HD;SILVERMAN, E
通讯作者：
SILVERMAN, E
影响因子：
7.8
作者：
HORNIK, K;STINCHCOMBE, M;WHITE, H
通讯作者：
WHITE, H