Liu Haipeng, Guan Xiao, Xu Beibei, Zeng Feiyue, Chen Changyong, Yin Hong Ling, Yi Xiaoping, Peng Yousong, Chen Bihong T
Department of Radiology, Xiangya Hospital, Central South University, Changsha, China.
National Clinical Research Center for Geriatric Disorders, Xiangya Hospital, Changsha, China.
Front Endocrinol (Lausanne). 2022 Mar 21;13:833413. doi: 10.3389/fendo.2022.833413. eCollection 2022.
To assess the accuracy of computed tomography (CT)-based machine learning models for differentiating subclinical pheochromocytoma (sPHEO) from lipid-poor adenoma (LPA) in patients with adrenal incidentalomas.
The study included 188 tumors in the 183 patients with LPA and 92 tumors in 86 patients with sPHEO. Pre-enhanced CT imaging features of the tumors were evaluated. Machine learning prediction models and scoring systems for differentiating sPHEO from LPA were built using logistic regression (LR), support vector machine (SVM) and random forest (RF) approaches.
The LR model performed better than other models. The LR model (M1) including three CT features: CT value, shape, and necrosis/cystic changes had an area under the receiver operating characteristic curve (AUC) of 0.917 and an accuracy of 0.864. The LR model (M2) including three CT features: CT value, shape and homogeneity had an AUC of 0.888 and an accuracy of 0.832. The S2 scoring system (sensitivity: 0.859, specificity: 0.824) had comparable diagnostic value to S1 (sensitivity: 0.815; specificity: 0.910).
Our results indicated the potential of using a non-invasive imaging method such as CT-based machine learning models and scoring systems for predicting histology of adrenal incidentalomas. This approach may assist the diagnosis and personalized care of patients with adrenal tumors.
评估基于计算机断层扫描（CT）的机器学习模型在鉴别肾上腺偶发瘤患者中亚临床嗜铬细胞瘤（sPHEO）与乏脂性腺瘤（LPA）方面的准确性。
该研究纳入了183例LPA患者的188个肿瘤以及86例sPHEO患者的92个肿瘤。对肿瘤的平扫CT影像特征进行评估。使用逻辑回归（LR）、支持向量机（SVM）和随机森林（RF）方法构建了用于区分sPHEO与LPA的机器学习预测模型和评分系统。
LR模型的表现优于其他模型。包含CT值、形态和坏死/囊性变这三个CT特征的LR模型（M1）的受试者操作特征曲线下面积（AUC）为0.917，准确率为0.864。包含CT值、形态和均匀性这三个CT特征的LR模型（M2）的AUC为0.888，准确率为0.832。S2评分系统（敏感性：0.859，特异性：0.824）与S1评分系统（敏感性：0.815；特异性：0.910）具有相当的诊断价值。
我们的结果表明，使用基于CT的机器学习模型和评分系统等非侵入性成像方法预测肾上腺偶发瘤组织学的潜力。这种方法可能有助于肾上腺肿瘤患者的诊断和个性化治疗。
