Department of Radiology, Zhongnan Hospital of Wuhan University, Wuhan, China.
Media Tek, Wuhan, China.
Med Phys. 2023 Apr;50(4):2049-2060. doi: 10.1002/mp.16177. Epub 2023 Jan 6.
Accurate diagnosis of N2 lymph node status of the resectable stage I-II non-small cell lung cancer (NSCLC) before surgery is crucial, while there is lack of corresponding method clinically.
To develop and validate a model to quantitively predict the N2 lymph node metastasis in presurgical clinical stage I-II NSCLC using multiview radiomics and deep learning method.
In this study, 140 NSCLC patients were enrolled and randomly divided into training and test sets. Univariate and multiple analysis method were used step by step to establish the clinical model; Then a multiview radiomics modeling scheme was designed, in which the optimal input feature set was determined by subcategorizing radiomics features (C1: original; C2: LoG and C3: wavelet) and comparison of corresponding radiomics model. The minimum-redundancy maximum-relevance (mRMR) selection and the least absolute shrinkage and selection operator (LASSO) algorithm were used for the feature selection and construction of each radiomics model (Rad). Next, an end-to-end ResNet18 architecture and transfer learning techniques were designed to construct a deep learning model (DL). Subsequently, the screened clinical risk factors and constructed Rad and DL models were combined and compared and a nomogram was constructed. Finally, the diagnostic performance of all constructed models were evaluated and compared using receiver operating characteristic curve (ROC) analysis, Delong test, Calibration analysis, Hosmer-Lemeshow test, and decision curves, respectively.
Carcinoma embryonic antigen (CEA) level and spiculation were screened to make up the Clinical model, while seven radiomics features in the optimal input feature set C2 + C3 were selected to construct the Rad. DL was constructed by training on 1.8 million natural images and small sample data of our N2 lymph node volume of interest (VOI) images. Except for the Clinical model, all other models showed good predictive accuracy and consistency in both training set and test set. DL (area under curve (AUC): 0.83) was better than Rad (AUC: 0.76) in predictive accuracy, but their difference was not significant (p = 0.45). The combined models showed better diagnostic performance than the model only clinical or image risk factors were used (AUC for Clinical, Rad + DL, Rad + Clinical, DL + Clinical, and Rad + DL + Clinical were respectively 0.66, 0.86, 0.82, 0.86, and 0.88). Finally, the Rad + DL + Clinical model with the best diagnostic performance was selected to draw the final nomogram for clinical use.
This study proposes a nomogram based on multiview radiomics, deep learning, and clinical features that can be efficiently used to quantitively predict presurgical N2 diseases in patients with clinical stage I-II NSCLC.
在手术前准确诊断可切除 I-II 期非小细胞肺癌（NSCLC）的 N2 淋巴结状态至关重要，但临床上缺乏相应的方法。
使用多视图放射组学和深度学习方法开发和验证一种模型，以定量预测术前临床 I-II 期 NSCLC 的 N2 淋巴结转移。
本研究纳入 140 例 NSCLC 患者，并随机分为训练集和测试集。使用单变量和多变量分析方法逐步建立临床模型；然后设计了一种多视图放射组学建模方案，通过对放射组学特征进行子分类（C1：原始；C2：LoG 和 C3：小波）和比较相应的放射组学模型来确定最佳输入特征集。最小冗余最大相关性（mRMR）选择和最小绝对值收缩和选择算子（LASSO）算法用于特征选择和每个放射组学模型（Rad）的构建。接下来，设计了一个端到端的 ResNet18 架构和迁移学习技术来构建深度学习模型（DL）。然后，筛选出的临床风险因素以及构建的 Rad 和 DL 模型进行组合和比较，并构建了一个列线图。最后，分别使用受试者工作特征曲线（ROC）分析、Delong 检验、校准分析、Hosmer-Lemeshow 检验和决策曲线对所有构建模型的诊断性能进行评估和比较。
筛选癌胚抗原（CEA）水平和分叶状来构建临床模型，而在最佳输入特征集 C2+C3 中选择了七个放射组学特征来构建 Rad。通过对 180 万张自然图像和我们的 N2 淋巴结感兴趣区（VOI）图像的小样本数据进行训练，构建了 DL。除临床模型外，所有其他模型在训练集和测试集均表现出良好的预测准确性和一致性。DL（曲线下面积（AUC）：0.83）在预测准确性方面优于 Rad（AUC：0.76），但差异无统计学意义（p=0.45）。与仅使用临床或图像风险因素的模型相比，联合模型显示出更好的诊断性能（临床、Rad+DL、Rad+临床、DL+临床和 Rad+DL+临床的 AUC 分别为 0.66、0.86、0.82、0.86 和 0.88）。最后，选择具有最佳诊断性能的 Rad+DL+临床模型绘制最终列线图供临床使用。
本研究提出了一种基于多视图放射组学、深度学习和临床特征的列线图，可以有效地用于定量预测临床 I-II 期 NSCLC 患者术前的 N2 疾病。
