基于 F-FDG-PET 的放射组学和深度学习特征的功效，采用机器学习方法预测胸腺瘤的病理风险亚型。
The efficacy of F-FDG-PET-based radiomic and deep-learning features using a machine-learning approach to predict the pathological risk subtypes of thymic epithelial tumors.
机构信息
Department of Radiology, Kagoshima University, Graduate School of Medical and Dental Sciences, Kagoshima, Japan.
Department of General Thoracic Surgery, Kagoshima University, Graduate School of Medical and Dental Sciences, Kagoshima, Japan.
出版信息
Br J Radiol. 2022 Jun 1;95(1134):20211050. doi: 10.1259/bjr.20211050. Epub 2022 Mar 28.
OBJECTIVE
To examine whether the machine-learning approach using 18-fludeoxyglucose positron emission tomography (F-FDG-PET)-based radiomic and deep-learning features is useful for predicting the pathological risk subtypes of thymic epithelial tumors (TETs).
METHODS
This retrospective study included 79 TET [27 low-risk thymomas (types A, AB and B1), 31 high-risk thymomas (types B2 and B3) and 21 thymic carcinomas] patients who underwent pre-therapeutic F-FDG-PET/CT. High-risk TETs (high-risk thymomas and thymic carcinomas) were 52 patients. The 107 PET-based radiomic features, including SUV-related parameters [maximum SUV (SUV), metabolic tumor volume (MTV), and total lesion glycolysis (TLG)] and 1024 deep-learning features extracted from the convolutional neural network were used to predict the pathological risk subtypes of TETs using six different machine-learning algorithms. The area under the curves (AUCs) were calculated to compare the predictive performances.
RESULTS
SUV-related parameters yielded the following AUCs for predicting thymic carcinomas: SUVmax 0.713, MTV 0.442, and TLG 0.479 or high-risk TETs: SUVmax 0.673, MTV 0.533, and TLG 0.539. The best-performing algorithm was the logistic regression model for predicting thymic carcinomas (AUC 0.900, accuracy 81.0%), and the random forest (RF) model for high-risk TETs (AUC 0.744, accuracy 72.2%). The AUC was significantly higher in the logistic regression model than three SUV-related parameters for predicting thymic carcinomas, and in the RF model than MTV and TLG for predicting high-risk TETs (each; < 0.05).
CONCLUSION
F-FDG-PET-based radiomic analysis using a machine-learning approach may be useful for predicting the pathological risk subtypes of TETs.
ADVANCES IN KNOWLEDGE
Machine-learning approach using F-FDG-PET-based radiomic features has the potential to predict the pathological risk subtypes of TETs.
目的
探讨基于 18 氟-脱氧葡萄糖正电子发射断层扫描（F-FDG-PET）的放射组学和深度学习特征的机器学习方法是否可用于预测胸腺瘤（TET）的病理风险亚型。
方法
本回顾性研究纳入 79 例接受治疗前 F-FDG-PET/CT 检查的 TET [27 例低危胸腺瘤（A型、AB 型和 B1 型）、31 例高危胸腺瘤（B2 型和 B3 型）和 21 例胸腺癌]患者。高危 TET（高危胸腺瘤和胸腺癌）患者 52 例。使用六种不同的机器学习算法，基于 107 个 PET 放射组学特征（包括 SUV 相关参数[最大 SUV（SUV）、代谢肿瘤体积（MTV）和总病变糖酵解（TLG）]和从卷积神经网络中提取的 1024 个深度学习特征）来预测 TET 的病理风险亚型。计算曲线下面积（AUCs）以比较预测性能。
结果
SUV 相关参数预测胸腺癌的 AUC 分别为：SUVmax 0.713、MTV 0.442 和 TLG 0.479，预测高危 TET 的 AUC 分别为：SUVmax 0.673、MTV 0.533 和 TLG 0.539。预测胸腺癌的最佳算法是逻辑回归模型（AUC 0.900，准确性 81.0%），预测高危 TET 的最佳算法是随机森林（RF）模型（AUC 0.744，准确性 72.2%）。与 SUV 相关参数相比，逻辑回归模型预测胸腺癌的 AUC 显著更高（均<0.05），与 MTV 和 TLG 相比，RF 模型预测高危 TET 的 AUC 更高（均<0.05）。
结论
基于 F-FDG-PET 的放射组学分析结合机器学习方法可能有助于预测 TET 的病理风险亚型。
知识进展
基于 F-FDG-PET 的放射组学特征的机器学习方法有可能预测 TET 的病理风险亚型。
Zotero 插件