Department of Radiology, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Shuaifuyuan No 1, Wangfujing Street, Dongcheng District, Beijing 100730, People's Republic of China (W.D., J.L., F.X., S.W., X.Z., Z.J., H.X.).
Department of Radiology, Beijing Tiantan Hospital, Capital Medical University, Fengtai District, Beijing, 100070, People's Republic of China (X.W.).
Acad Radiol. 2024 May;31(5):1889-1897. doi: 10.1016/j.acra.2023.09.043. Epub 2023 Nov 17.
According to current guidelines, pancreatic cystic lesions (PCLs) with worrisome or high-risk features may have overtreatment. The purpose of this study was to build a clinical and radiological based machine-learning (ML) model to identify malignant PCLs for surgery among preoperative PCLs with worrisome or high-risk features.
Clinical and radiological details of 317 pathologically confirmed PCLs with worrisome or high-risk features were retrospectively analyzed and applied to ML models including Support Vector Machine, Logistic Regression (LR), Decision Tree, Bernoulli NB, Gaussian NB, K Nearest Neighbors and Linear Discriminant Analysis. The diagnostic ability for malignancy of the optimal model with the highest diagnostic AUC in the cross-validation procedure was further evaluated in internal (n = 77) and external (n = 50) testing cohorts, and was compared to two published guidelines in internal mucinous cyst cohort.
Ten clinical and radiological feature-based LR model was the optimal model with the highest AUC (0.951) in the cross-validation procedure. In the internal testing cohort, LR model reached an AUC, accuracy, sensitivity, and specificity of 0.927, 0.909, 0.914, and 0.905; in the external testing cohort, LR model reached 0.948, 0.900, 0.963, and 0.826. When compared to the European guidelines and the ACG guidelines, LR model demonstrated significantly better accuracy and specificity in identifying malignancy, while maintaining the same high sensitivity.
Clinical- and radiological-based LR model can accurately identify malignant PCLs in patients with worrisome or high-risk features, possessing diagnostic performance better than the European guidelines as well as ACG guidelines.
根据现有指南，具有令人担忧或高危特征的胰腺囊性病变（PCL）可能存在过度治疗的情况。本研究旨在构建一种基于临床和影像学的机器学习（ML）模型，以识别术前具有令人担忧或高危特征的 PCL 中需要手术的恶性 PCL。
回顾性分析了 317 例经病理证实的具有令人担忧或高危特征的 PCL 的临床和影像学资料，并将其应用于包括支持向量机、逻辑回归（LR）、决策树、伯努利 NB、高斯 NB、K 近邻和线性判别分析在内的 ML 模型中。在交叉验证过程中，最优模型的诊断效能（最高 AUC）用于评估内部（n=77）和外部（n=50）测试队列中的恶性肿瘤诊断能力，并与内部黏液性囊腺癌队列中的两个已发表的指南进行比较。
基于 10 项临床和影像学特征的 LR 模型是交叉验证过程中 AUC 最高（0.951）的最优模型。在内部测试队列中，LR 模型的 AUC、准确性、敏感性和特异性分别为 0.927、0.909、0.914 和 0.905；在外部测试队列中，LR 模型分别为 0.948、0.900、0.963 和 0.826。与欧洲指南和 ACG 指南相比，LR 模型在识别恶性肿瘤方面具有更高的准确性和特异性，同时保持了相同的高敏感性。
基于临床和影像学的 LR 模型可以准确识别具有令人担忧或高危特征的患者中的恶性 PCL，其诊断性能优于欧洲指南和 ACG 指南。
