? 基于多模态特征和多分类器融合的前列腺癌放疗中直肠并发症预测模型 365bet官方网投_365bet足球比分_365bet足彩
 文章快速检索 ? ? 高级检索
 ??南方医科大学学报??2019, Vol. 39Issue (8): 972-979??DOI: 10.12122/j.issn.1673-4254.2019.08.15. 0

### 引用本文?[复制中英文]

HE Qiang, WANG Xuetao, LI Xin, ZHEN Xin. Prediction of rectal toxicity of radiotherapy for prostate cancer based on multi-modality feature and multi-classifiers[J]. Journal of Southern Medical University, 2019, 39(8): 972-979. DOI: 10.12122/j.issn.1673-4254.2019.08.15.

### 文章历史

1. 南方医科大学 生物医学工程学院，广东 广州 510515;
2. 广州中医药大学第二附属医院，广东 广州 510120

Prediction of rectal toxicity of radiotherapy for prostate cancer based on multi-modality feature and multi-classifiers
HE Qiang 1, WANG Xuetao 2, LI Xin 1, ZHEN Xin 1 ????
1. School of Biomedical Engineering, Southern Medical University, Guangzhou 510515, China;
2. Second Affiliated Hospital of Guangzhou University of Chinese Medicine, Guangzhou, 510120, China
Supported by the National Natural Science Foundation of China (81728016 and 81874216), the National Key Research and Development Program of China (2017YFC0112900)
Abstract: Objective To evaluate rectal toxicity of radiotherapy for prostate cancer using a novel predictive model based on multi-modality and multi-classifier fusion. Methods We retrospectively collected the clinical data from 44 prostate cancer patients receiving external beam radiation (EBRT), including the treatment data, clinical parameters, planning CT data and the treatment plans. The clinical parameter features and dosimetric features were extracted as two different modality features, and a subset of features was selected to train the 5 base classifiers (SVM, Decision Tree, K-nearest-neighbor, Random forests and XGBoost). To establish the multi-modality and multi-classifier fusion model, a multi-criteria decision-making based weight assignment algorithm was used to assign weights for each base classifier under the same modality. A repeat 5-fold cross-validation and the 4 indexes including the area under ROC curve (AUC), accuracy, sensitivity and specificity were used to evaluate the proposed model. In addition, the proposed model was compared quantitatively with different feature selection methods, different weight allocation algorithms, the model based on single mode single classifier, and two integrated models using other fusion methods. Results Repeated (5 times) 5-fold cross validation of the proposed model showed an accuracy of 0.78 for distinguishing toxicity from non-toxicity with an AUC of 0.83, a specificity of 0.79 and a sensitivity of 0.76. Conclusion Compared with the models based on a single mode or a single classifier and other fusion models, the proposed model can more accurately predict rectal toxicity of radiotherapy for prostate cancer.
Keywords: multi-modality????multi-classifier????multi-criteria decision-making????prostate cancer radiotherapy????rectal toxicity????

1 资料和方法 1.1 研究对象

1.2 方法概述

 图 1 模型框架示意图 Fig.1 Framework of the proposed model.
1.2.1 特征提取

1.2.2 特征选择

1.2.3 基于多准则决策的权重分配算法

1.2.4 多模态特征及多分类器融合过程

1.3 模型验证和评估 1.3.1 模型验证

（1）特征选择方法的验证

（2）基于多准则决策的权重分配算法的验证

 ${\rm{WF}}1:\quad {w_n} = \frac{{{\eta _n}}}{{\sum\limits_i {{\eta _i}} }},$ (1)
 ${\rm{WF}}2:\quad {w_n} = \frac{{{\eta _n} - {\eta _l}}}{{{\eta _u} - {\eta _l}}},$ (2)
 ${\rm{WF}}3:\quad {w_n} = \log \frac{{{\eta _n}}}{{1 - {\eta _n}}},$ (3)

（3）与单模态单分类器模型的比较

（4）与不同融合模型的比较

 图 2 对比模型S1的融合示意图 Fig.2 Comparison reference ensemble scheme 1 (S1) to aggregate multi-modality features using Random Forests or XGBoost.
 图 3 对比模型S2的融合示意图 Fig.3 Comparison reference ensemble scheme 2 (S2) to aggregate multi-modality features using PV, WAF or Stacking.
1.3.2 评价指标

Accuracy=(TP+TN)/(TP+FP+FN+TN)

Specificity=TN/(TN+FP)

Sensitivity=TP/(TP+FN)，

2 结果 2.1 特征选择方法的验证

2.2 基于多准则决策的权重分配算法的验证

2.3 与单模态单分类器的比较

2.4 与不同融合模型的比较

 图 4 本研究提出的模型与S1、S2的ROC曲线 Fig.4 ROC analysis between H-MCF, S1, and S2.
3 讨论