Hou Liqiao, Chen Kuifei, Zhou Chao, Tang Xingni, Yu Changhui, Jia Haijian, Xu Qianyi, Zhou Suna, Yang Haihua
Key Laboratory of Radiation Oncology of Taizhou, Radiation Oncology Institute of Enze Medical Health Academy , Department of Radiation Oncology, Taizhou Hospital of Zhejiang Province Affiliated to Wenzhou Medical University, NO.150 Ximen Street, Linhai, Taizhou City, 317000, Zhejiang Province, China.
Department of Radiation Oncology, Enze Hospital Affiliated Hospital of Hangzhou Medical College, Zhejiang Province 317000, China.
Clin Transl Radiat Oncol. 2024 Jul 31;48:100828. doi: 10.1016/j.ctro.2024.100828. eCollection 2024 Sep.
To establish a radiomics model using radiomics features from different region of interests (ROI) based on dosimetry-related regions in enhanced computed tomography (CT) simulated images to predict radiation pneumonitis (RP) in patients with non-small cell lung cancer (NSCLC).
Our retrospective study was conducted based on a cohort of 236 NSCLC patients (59 of them with RP≥2) who were treated in 2 institutions and divided into the primary cohort (n = 182,46 of them with RP≥2) and external validation cohort (n = 54,13 of them with RP≥2). Radiomic features extracted from three ROIs were defined as the whole lung (WL), the dose volume histogram (DVH) of the lung V20 (V20_Lung) and the DVH of the V30 of lung minus the planning target volume (PTV) (V30 Lung-PTV). A total of 107 radiomics features were extracted from each ROIs. The test, correlation coefficient and least absolute shrinkage and selection operator (LASSO) were performed for features selection. Six models based on different classification algorithms were developed to select the best radiomics model (R model).In addition, we built a dosimetry model then combined it with the best R model to create a mixed model (R+D model) The receiver operating characteristic (ROC) curve was delineated to assess the predictive efficacy of the models. Decision curve analysis could benefit from the model proposals through the assessment of clinical utility.
Among the three ROIs, the best R model constructed from the LightGBM algorithm demonstrated the strongest discriminative ability in the ROI of V30 Lung-PTV. The corresponding area under the curve (AUC) value was 0.930 (95 % confidence interval (CI): 0.829-0.941). The D model, R model and R+D model achieved AUC values of 0.798 (95 %CI: 0.732-0.865), 0.930 (95 %CI: 0.829-0.941) and 0.940 (95 %CI: 0.906-0.974) in primary cohort, and in external validation cohort, the AUC values were 0.793 (95 %CI:0.637-0.949), 0.887 (95 %CI:0.810-0.993), 0.951 (95CI%:0.891-1.000). Decision curve demonstrate that R+D model could benefit for patients through the assessment of clinical utility.
The radiomics model was able to predict the acute RP more effectively in comparison with the traditional dosimetry model. Especially the radiomics model based on the V30 Lung-PTV region was able to achieve a higher accuracy when compared to the other regions.
基于增强计算机断层扫描（CT）模拟图像中与剂量学相关的区域，利用来自不同感兴趣区域（ROI）的放射组学特征建立放射组学模型，以预测非小细胞肺癌（NSCLC）患者的放射性肺炎（RP）。
我们的回顾性研究基于236例在2家机构接受治疗的NSCLC患者队列（其中59例RP≥2级）进行，分为初级队列（n = 182，其中46例RP≥2级）和外部验证队列（n = 54，其中13例RP≥2级）。从三个ROI提取的放射组学特征定义为全肺（WL）、肺V20的剂量体积直方图（DVH）（V20_Lung）以及肺V30减去计划靶体积（PTV）的DVH（V30 Lung-PTV）。从每个ROI中总共提取了107个放射组学特征。对特征进行检验、相关系数分析以及最小绝对收缩和选择算子（LASSO）分析以进行特征选择。基于不同分类算法开发了六个模型以选择最佳放射组学模型（R模型）。此外，我们构建了一个剂量学模型，然后将其与最佳R模型相结合以创建一个混合模型（R+D模型）。绘制受试者操作特征（ROC）曲线以评估模型的预测效能。决策曲线分析可通过评估临床实用性从模型建议中获益。
在三个ROI中，由LightGBM算法构建的最佳R模型在V30 Lung-PTV的ROI中表现出最强的判别能力。相应的曲线下面积（AUC）值为0.930（95%置信区间（CI）：0.829 - 0.941）。在初级队列中，D模型、R模型和R+D模型的AUC值分别为0.798（95%CI：0.732 - 0.865）、0.930（95%CI：0.829 - 0.941）和0.940（95%CI：0.906 - 0.974），在外部验证队列中，AUC值分别为0.793（95%CI：0.637 - 0.949）、0.887（95%CI：0.810 - 0.993）、0.951（95%CI：0.891 - 1.000）。决策曲线表明R+D模型通过临床实用性评估可为患者带来益处。
与传统剂量学模型相比，放射组学模型能够更有效地预测急性RP。特别是基于V30 Lung-PTV区域的放射组学模型与其他区域相比能够实现更高的准确性。
