Wu Jian, Lian Chunfeng, Ruan Su, Mazur Thomas R, Mutic Sasa, Anastasio Mark A, Grigsby Perry W, Vera Pierre, Li Hua
Department of Radiation Oncology, Washington University, Saint louis, MO 63110 USA.
Laboratoire LITIS (EA 4108), Equipe Quantif, University of Rouen, France.
IEEE Trans Radiat Plasma Med Sci. 2019 Mar;3(2):216-224. doi: 10.1109/TRPMS.2018.2872406. Epub 2018 Sep 27.
In this study, we proposed a new radiomics-based treatment outcome prediction model for cancer patients. The prediction model is developed based on belief function theory (BFT) and sparsity learning to address the challenges of redundancy, heterogeneity, and uncertainty of radiomic features, and relatively small-sized and unbalanced training samples. The model first selects the most predictive feature subsets from relatively large amounts of radiomic features extracted from pre- and/or in-treatment positron emission tomography (PET) images and available clinical and demographic features. Then an evidential k-nearest neighbor (EK-NN) classifier is proposed to utilize the selected features for treatment outcome prediction. Twenty-five stage II-III lung, 36 esophagus, 63 stage II-III cervix, and 45 lymphoma cancer patient cases were included in this retrospective study. Performance and robustness of the proposed model were assessed with measures of feature selection stability, outcome prediction accuracy, and receiver operating characteristics (ROC) analysis. Comparison with other methods were conducted to demonstrate the feasibility and superior performance of the proposed model.
在本研究中，我们为癌症患者提出了一种基于放射组学的新治疗结果预测模型。该预测模型基于信念函数理论（BFT）和稀疏学习开发，以应对放射组学特征的冗余性、异质性和不确定性，以及相对较小规模和不平衡的训练样本等挑战。该模型首先从治疗前和/或治疗中的正电子发射断层扫描（PET）图像中提取的大量放射组学特征以及可用的临床和人口统计学特征中选择最具预测性的特征子集。然后提出了一种证据k近邻（EK-NN）分类器，利用所选特征进行治疗结果预测。这项回顾性研究纳入了25例II-III期肺癌、36例食管癌、63例II-III期宫颈癌和45例淋巴瘤癌患者病例。通过特征选择稳定性、结果预测准确性和受试者操作特征（ROC）分析等指标评估了所提出模型的性能和稳健性。与其他方法进行了比较，以证明所提出模型的可行性和优越性能。
