基于2.5D深度学习的使用增强CT预测透明细胞肾细胞癌病理分级：一项多中心研究

2.5D Deep Learning-Based Prediction of Pathological Grading of Clear Cell Renal Cell Carcinoma Using Contrast-Enhanced CT: A Multicenter Study.

作者信息

Yang Zi, Jiang Haitao, Shan Shuai, Wang Xu, Kou Quanming, Wang Chao, Jin Pengfei, Xu Yuyun, Liu Xiaohui, Zhang Yudong, Zhang Yuqing

机构信息

School of Automation, Hangzhou Dianzi University, Hangzhou 310018, P.R. China (Z.Y., Q.K., Y.Z.).

Department of Radiology, Zhejiang Cancer Hospital, Hangzhou 310022, China (H.J., X.W., P.J.).

出版信息

Acad Radiol. 2025 Jul 19. doi: 10.1016/j.acra.2025.06.056.

DOI:10.1016/j.acra.2025.06.056

PMID:40683765

Abstract

RATIONALE AND OBJECTIVES

To develop and validate a deep learning model based on arterial phase-enhanced CT for predicting the pathological grading of clear cell renal cell carcinoma (ccRCC).

MATERIALS AND METHODS

Data from 564 patients diagnosed with ccRCC from five distinct hospitals were retrospectively analyzed. Patients from centers 1 and 2 were randomly divided into a training set (n=283) and an internal test set (n=122). Patients from centers 3, 4, and 5 served as external validation sets 1 (n=60), 2 (n=38), and 3 (n=61), respectively. A 2D model, a 2.5D model (three-slice input), and a radiomics-based multi-layer perceptron (MLP) model were developed. Model performance was evaluated using the area under the curve (AUC), accuracy, and sensitivity.

RESULTS

The 2.5D model outperformed the 2D and MLP models. Its AUCs were 0.959 (95% CI: 0.9438-0.9738) for the training set, 0.879 (95% CI: 0.8401-0.9180) for the internal test set, and 0.870 (95% CI: 0.8076-0.9334), 0.862 (95% CI: 0.7581-0.9658), and 0.849 (95% CI: 0.7766-0.9216) for the three external validation sets, respectively. The corresponding accuracy values were 0.895, 0.836, 0.827, 0.825, and 0.839. Compared to the MLP model, the 2.5D model achieved significantly higher AUCs (increases of 0.150 [p<0.05], 0.112 [p<0.05], and 0.088 [p<0.05]) and accuracies (increases of 0.077 [p<0.05], 0.075 [p<0.05], and 0.101 [p<0.05]) in the external validation sets.

CONCLUSION

The 2.5D model based on 2.5D CT image input demonstrated improved predictive performance for the WHO/ISUP grading of ccRCC.

摘要

原理与目的

开发并验证一种基于动脉期增强CT的深度学习模型，用于预测透明细胞肾细胞癌（ccRCC）的病理分级。

材料与方法

回顾性分析来自五家不同医院的564例诊断为ccRCC的患者的数据。来自中心1和中心2的患者被随机分为训练集（n = 283）和内部测试集（n = 122）。来自中心3、4和5的患者分别作为外部验证集1（n = 60）、2（n = 38）和3（n = 61）。开发了一个二维模型、一个2.5D模型（三层切片输入）和一个基于影像组学的多层感知器（MLP）模型。使用曲线下面积（AUC）、准确率和灵敏度评估模型性能。

结果

2.5D模型优于二维模型和MLP模型。其在训练集的AUC为0.959（95%CI：0.9438 - 0.9738），在内部测试集为0.879（95%CI：0.8401 - 0.9180），在三个外部验证集分别为0.870（95%CI：0.8076 - 0.9334）、0.862（95%CI：0.7581 - 0.9658）和0.849（95%CI：0.7766 - 0.9216）。相应的准确率值分别为0.895、0.836、0.827、0.825和0.839。与MLP模型相比，2.5D模型在外部验证集中的AUC显著更高（分别增加0.150 [p < 0.05]、0.112 [p < 0.05]和0.088 [p < 0.05]），准确率也显著更高（分别增加0.077 [p < 0.05]、0.075 [p < 0.05]和0.101 [p < 0.05]）。