A survival prediction model via interpretable machine learning for patients with oropharyngeal cancer following radiotherapy
A survival prediction model via interpretable machine learning for patients with oropharyngeal cancer following radiotherapy
复制标题
DOI：
10.1007/s00432-023-04644-y
复制
发表时间：
2023-02-18
影响因子：
3.6
通讯作者：
Qi, X. Sharon
中科院分区：
文献类型：
--
作者：
Pan, Xiaoying;Feng, Tianhao;Qi, X. Sharon
目的探索可解释的机器学习（ML）方法，以期为口咽癌（OPC）患者的生存预测增加更多的预后价值。方法对来自 TCIA 数据库的 427 例 OPC 患者（训练 341，测试 86）进行分析。使用 Pyradiomics 从计划 CT 中提取的大体肿瘤体积 (GTV) 的放射组学特征以及 HPV p16 状态等患者特征被认为是潜在的预测因素。提出了一种由最小绝对选择算子（Lasso）和顺序浮动向后选择（SFBS）组成的多级降维算法，以有效去除冗余/不相关的特征。通过 Shapley-Additive-exPlanations (SHAP) 算法量化每个特征对 Extreme-Gradient-Boosting (XGBoost) 决策的贡献来构建可解释模型。结果本研究提出的 Lasso-SFBS 算法最终选择了 14 个特征，我们的预测模型在基于该特征集的测试数据集上实现了 0.85 的 ROC 曲线下面积 (AUC)。 SHAP 计算的贡献值排序显示，与生存最相关的预测因子是 ECOG 表现状态、wavelet-LLH_firstorder_Mean、化疗、wavelet-LHL_glcm_InverseVariance、肿瘤大小。那些接受化疗、HPV p16 状态呈阳性且 ECOG 表现状态较低的患者往往具有较高的 SHAP 评分和较长的生存期；诊断时年龄较大、有大量饮酒和吸烟史的患者往往 SHAP 评分较低且生存期较短。结论我们证明了患者特征和影像学特征相结合对 OPC 患者总生存期的预测价值。多级降维算法可以可靠地识别与总体生存率最相关的最合理的预测因子。开发可解释的患者特异性生存预测模型，捕获每个预测因素与临床结果的相关性，以促进个性化治疗的临床决策。
PurposeTo explore interpretable machine learning (ML) methods, with the hope of adding more prognosis value, for predicting survival for patients with Oropharyngeal-Cancer (OPC).MethodsA cohort of 427 OPC patients (Training 341, Test 86) from TCIA database was analyzed. Radiomic features of gross-tumor-volume (GTV) extracted from planning CT using Pyradiomics, and HPV p16 status, etc. patient characteristics were considered as potential predictors. A multi-level dimension reduction algorithm consisting of Least-Absolute-Selection-Operator (Lasso) and Sequential-Floating-Backward-Selection (SFBS) was proposed to effectively remove redundant/irrelevant features. The interpretable model was constructed by quantifying the contribution of each feature to the Extreme-Gradient-Boosting (XGBoost) decision by Shapley-Additive-exPlanations (SHAP) algorithm.ResultsThe Lasso-SFBS algorithm proposed in this study finally selected 14 features, and our prediction model achieved an area-under-ROC-curve (AUC) of 0.85 on the test dataset based on this feature set. The ranking of the contribution values calculated by SHAP shows that the top predictors that were most correlated with survival were ECOG performance status, wavelet-LLH_firstorder_Mean, chemotherapy, wavelet-LHL_glcm_InverseVariance, tumor size. Those patients who had chemotherapy, with positive HPV p16 status, and lower ECOG performance status, tended to have higher SHAP scores and longer survival; who had an older age at diagnosis, heavy drinking and smoking pack year history, tended to lower SHAP scores and shorter survival.ConclusionWe demonstrated predictive values of combined patient characteristics and imaging features for the overall survival of OPC patients. The multi-level dimension reduction algorithm can reliably identify the most plausible predictors that are mostly associated with overall survival. The interpretable patient-specific survival prediction model, capturing correlations of each predictor and clinical outcome, was developed to facilitate clinical decision-making for personalized treatment.