Department of Radiology, Kagoshima University, Graduate School of Medical and Dental Sciences, 8-35-1 Sakuragaoka, Kagoshima, 890-8544, Japan.
Department of Management Planning Division, Harada Academy, 2-54-4 Higashitaniyama, Kagoshima, 890-0113, Japan.
Abdom Radiol (NY). 2022 Feb;47(2):838-847. doi: 10.1007/s00261-021-03350-y. Epub 2021 Nov 25.
To examine the usefulness of machine learning to predict prognosis in cervical cancer using clinical and radiomic features of 2-deoxy-2-[F]fluoro-D-glucose (F-FDG) positron emission tomography/computed tomography (CT) (F-FDG-PET/CT).
This retrospective study included 50 cervical cancer patients who underwent F-FDG-PET/CT before treatment. Four clinical (age, histology, stage, and treatment) and 41 F-FDG-PET-based radiomic features were ranked and a subset of useful features for association with disease progression was selected based on decrease of the Gini impurity. Six machine learning algorithms (random forest, neural network, k-nearest neighbors, naive Bayes, logistic regression, and support vector machine) were compared using the areas under the receiver operating characteristic curve (AUC). Progression-free survival (PFS) was assessed using Cox regression analysis.
The five top predictors of disease progression were: stage, surface area, metabolic tumor volume, gray-level run length non-uniformity (GLRLM_RLNU), and gray-level non-uniformity for run (GLRLM_GLNU). The naive Bayes model was the best-performing classifier for predicting disease progression (AUC = 0.872, accuracy = 0.780, F1 score = 0.781, precision = 0.788, and recall = 0.780). In the naive Bayes model, 5-year PFS was significantly higher in predicted non-progression than predicted progression (80.1% vs. 9.1%, p < 0.001) and was only the independent factor for PFS in multivariate analysis (HR, 6.89; 95% CI, 1.92-24.69; p = 0.003).
A machine learning approach based on clinical and pretreatment F-FDG PET-based radiomic features may be useful for predicting tumor progression in cervical cancer patients.
利用 2-脱氧-2-[F]氟代-D-葡萄糖（F-FDG）正电子发射断层扫描/计算机断层扫描（PET/CT）的临床和放射组学特征，探讨机器学习在预测宫颈癌预后中的应用价值。
本回顾性研究纳入了 50 例在治疗前接受 F-FDG-PET/CT 检查的宫颈癌患者。共提取了 4 个临床（年龄、组织学、分期和治疗）和 41 个 F-FDG-PET 基于的放射组学特征，并根据基尼不纯度的降低选择了一组与疾病进展相关的有用特征子集。使用受试者工作特征曲线（ROC）下面积（AUC）比较了 6 种机器学习算法（随机森林、神经网络、k-近邻、朴素贝叶斯、逻辑回归和支持向量机）。使用 Cox 回归分析评估无进展生存期（PFS）。
疾病进展的 5 个最重要预测因子为：分期、表面积、代谢肿瘤体积、灰度游程长度非均匀性（GLRLM_RLNU）和灰度游程非均匀性（GLRLM_GLNU）。朴素贝叶斯模型是预测疾病进展的最佳分类器（AUC=0.872，准确率=0.780，F1 评分=0.781，精确率=0.788，召回率=0.780）。在朴素贝叶斯模型中，预测非进展组的 5 年 PFS 显著高于预测进展组（80.1%比 9.1%，p<0.001），且在多变量分析中仅为 PFS 的独立因素（HR，6.89；95%CI，1.92-24.69；p=0.003）。
基于临床和治疗前 F-FDG PET 基于的放射组学特征的机器学习方法可能有助于预测宫颈癌患者的肿瘤进展。
