From the Departments of Radiology (Y.B., X.F., M.Z., J. Yu, H.Z., J.L., C.S.) and Pathology (H.J.), Changhai Hospital, 168 Changhai Road, Shanghai 200433, China; Ping An Technology, Shanghai, China (Z.Z.); and PAII Inc, Bethesda, Md (L.Z., J. Yao, L.L.).
Radiology. 2023 Jan;306(1):160-169. doi: 10.1148/radiol.220329. Epub 2022 Sep 6.
Background Although deep learning has brought revolutionary changes in health care, reliance on manually selected cross-sectional images and segmentation remain methodological barriers. Purpose To develop and validate an automated preoperative artificial intelligence (AI) algorithm for tumor and lymph node (LN) segmentation with CT imaging for prediction of LN metastasis in patients with pancreatic ductal adenocarcinoma (PDAC). Materials and Methods In this retrospective study, patients with surgically resected, pathologically confirmed PDAC underwent multidetector CT from January 2015 to April 2020. Three models were developed, including an AI model, a clinical model, and a radiomics model. CT-determined LN metastasis was diagnosed by radiologists. Multivariable logistic regression analysis was conducted to develop the clinical and radiomics models. The performance of the models was determined on the basis of their discrimination and clinical utility. Kaplan-Meier curves, the log-rank test, or Cox regression were used for survival analysis. Results Overall, 734 patients (mean age, 62 years ± 9 [SD]; 453 men) were evaluated. All patients were split into training ( = 545) and validation ( = 189) sets. Patients who had LN metastasis (LN-positive group) accounted for 340 of 734 (46%) patients. In the training set, the AI model showed the highest performance (area under the receiver operating characteristic curve [AUC], 0.91) in the prediction of LN metastasis, whereas the radiologists and the clinical and radiomics models had AUCs of 0.58, 0.76, and 0.71, respectively. In the validation set, the AI model showed the highest performance (AUC, 0.92) in the prediction of LN metastasis, whereas the radiologists and the clinical and radiomics models had AUCs of 0.65, 0.77, and 0.68, respectively ( < .001). AI model-predicted positive LN metastasis was associated with worse survival (hazard ratio, 1.46; 95% CI: 1.13, 1.89; = .004). Conclusion An artificial intelligence model outperformed radiologists and clinical and radiomics models for prediction of lymph node metastasis at CT in patients with pancreatic ductal adenocarcinoma. © RSNA, 2022 See also the editorial by Chu and Fishman in this issue.
背景 深度学习在医疗保健领域带来了革命性的变化，但对人工选择的横断面图像和分割的依赖仍然是方法学上的障碍。目的 开发和验证一种基于 CT 成像的术前人工智能（AI）算法，用于预测胰腺导管腺癌（PDAC）患者的肿瘤和淋巴结（LN）的分割，以预测 LN 转移。材料与方法 本回顾性研究纳入了 2015 年 1 月至 2020 年 4 月期间接受手术切除、病理证实的 PDAC 患者，共进行了多排 CT 检查。建立了三个模型，包括 AI 模型、临床模型和放射组学模型。由放射科医生通过 CT 确定 LN 转移情况。采用多变量逻辑回归分析建立临床和放射组学模型。基于判别能力和临床实用性来确定模型的性能。采用 Kaplan-Meier 曲线、对数秩检验或 Cox 回归进行生存分析。结果 共有 734 例患者（平均年龄 62 岁±9[标准差]；453 例男性）入组。所有患者被分为训练集（n=545）和验证集（n=189）。有 LN 转移（LN 阳性组）的患者占 734 例患者的 340 例（46%）。在训练集中，AI 模型在预测 LN 转移方面表现最佳（受试者工作特征曲线下面积[AUC]，0.91），而放射科医生和临床及放射组学模型的 AUC 分别为 0.58、0.76 和 0.71。在验证集中，AI 模型在预测 LN 转移方面表现最佳（AUC，0.92），而放射科医生和临床及放射组学模型的 AUC 分别为 0.65、0.77 和 0.68（<.001）。AI 模型预测的阳性 LN 转移与较差的生存相关（风险比，1.46；95%CI：1.13，1.89；<.001）。结论 在预测胰腺导管腺癌患者 CT 上的 LN 转移方面，人工智能模型优于放射科医生和临床及放射组学模型。
