Histologic subtype classification of non-small cell lung cancer using PET/CT images
Histologic subtype classification of non-small cell lung cancer using PET/CT images
复制标题
使用 PET/CT 图像对非小细胞肺癌进行组织学亚型分类
DOI：
10.1007/s00259-020-04771-5
复制
发表时间：
2020-08-10
影响因子：
9.1
通讯作者：
Guo, Xiuhua
中科院分区：
文献类型：
--
作者：
Han, Yong;Ma, Yuan;Guo, Xiuhua
目的评估PET/CT图像鉴别非小细胞肺癌（NSCLC）组织学亚型的能力，并从基于放射组学的机器学习/深度学习算法中确定最佳模型。方法回顾性分析867例腺癌（ADC）和552例鳞状细胞癌（SCC）的临床资料。采用分层随机样本283例（20%）作为测试集（173例ADC和110例SCC）；剩余数据作为训练集。每个肿瘤区域共提取688个特征。评估10种特征选择技术、10种机器学习（ML）模型和VGG16深度学习（DL）算法，构建ADC和SCC鉴别诊断的最佳分类模型。采用十倍交叉验证和网格搜索技术对训练数据集上的模型超参数进行评估和优化。使用受试者工作特征曲线下面积（AUROC）、准确度、精密度、灵敏度和特异性来评估模型在测试数据集上的性能。结果每种特征选择技术选出50个排名靠前的子集特征进行分类。线性判别分析（LDA） （AUROC, 0.863；准确率，0.794）和支持向量机（SVM） （AUROC, 0.863；准确率，0.792）两种分类器结合了1,2,1NR特征选择方法，均取得了最优的分类效果。随机森林（random forest， RF）分类器（AUROC, 0.824，准确率，0.775）和ld2,1nr特征选择方法（AUROC, 0.815，准确率，0.764）在我们所采用的分类器和特征选择方法中分别表现出优异的平均性能。此外，VGG16深度学习算法（AUROC, 0.903；准确率，0.841）在结合放射组学的情况下优于所有传统的机器学习方法。结论应用放射学机器学习/深度学习算法可以帮助放射科医生通过PET/CT图像区分NSCLC的组织学亚型。
PurposesTo evaluate the capability of PET/CT images for differentiating the histologic subtypes of non-small cell lung cancer (NSCLC) and to identify the optimal model from radiomics-based machine learning/deep learning algorithms.MethodsIn this study, 867 patients with adenocarcinoma (ADC) and 552 patients with squamous cell carcinoma (SCC) were retrospectively analysed. A stratified random sample of 283 patients (20%) was used as the testing set (173 ADC and 110 SCC); the remaining data were used as the training set. A total of 688 features were extracted from each outlined tumour region. Ten feature selection techniques, ten machine learning (ML) models and the VGG16 deep learning (DL) algorithm were evaluated to construct an optimal classification model for the differential diagnosis of ADC and SCC. Tenfold cross-validation and grid search technique were employed to evaluate and optimize the model hyperparameters on the training dataset. The area under the receiver operating characteristic curve (AUROC), accuracy, precision, sensitivity and specificity was used to evaluate the performance of the models on the test dataset.ResultsFifty top-ranked subset features were selected by each feature selection technique for classification. The linear discriminant analysis (LDA) (AUROC, 0.863; accuracy, 0.794) and support vector machine (SVM) (AUROC, 0.863; accuracy, 0.792) classifiers, both of which coupled with theℓ2,1NR feature selection method, achieved optimal performance. The random forest (RF) classifier (AUROC, 0.824; accuracy, 0.775) andℓ2,1NR feature selection method (AUROC, 0.815; accuracy, 0.764) showed excellent average performance among the classifiers and feature selection methods employed in our study, respectively. Furthermore, the VGG16 DL algorithm (AUROC, 0.903; accuracy, 0.841) outperformed all conventional machine learning methods in combination with radiomics.ConclusionEmploying radiomic machine learning/deep learning algorithms could help radiologists to differentiate the histologic subtypes of NSCLC via PET/CT images.