Nie Pei, Zhao Xia, Ma Jinlong, Wang Yicong, Li Ben, Li Xiaoli, Li Qiyuan, Wang Yanmei, Xu Yuchao, Dai Zhengjun, Wu Jie, Wang Ning, Yang Guangjie, Hao Dapeng, Yu Tengbo
Department of Radiology, the Affiliated Hospital of Qingdao University, No. 16, Jiangsu Road, Qingdao 266003, Shandong, China.
Department of Radiology, the Affiliated Hospital of Shandong University of Traditional Chinese Medicine, Jinan, Shandong, China.
Eur J Radiol. 2024 Dec;181:111719. doi: 10.1016/j.ejrad.2024.111719. Epub 2024 Sep 17.
Computed tomography (CT) and biopsy may be insufficient for preoperative evaluation of the grade and outcome of patients with chondrosarcoma. The aim of this study was to develop and validate a CT-based deep learning radiomics model (DLRM) for predicting histologic grade and prognosis in chondrosarcoma (CS).
A multicenter 211 (training cohort/ test cohort, 127/84) CS patients were enrolled. Radiomics signature (RS), deep learning signature (DLS), and DLRM incorporating radiomics and deep learning features were developed for predicting the grade. Kaplan-Meier survival analysis was used to assess the association of the model-predicted grade with recurrence-free survival (RFS). Model performance was evaluated with the area under the receiver operating characteristic curve (AUC) and the Harrell's concordance index (C-index).
The DLRM (AUC, 0.879; 95 % confidence interval [CI], 0.802-0.956) outperformed (z = 2.773, P=0.006) the RS (AUC, 0.715;95 % CI, 0.606-0.825) in predicting grade in the test cohort. RFS showed significant differences (log-rank test, P<0.05) between low-grade and high-grade patients stratified by DLRM. The DLRM achieved a higher C-index (0.805; 95 % CI, 0.694-0.916) than the RS (0.692, 95 % CI, 0.540-0.844) did in predicting RFS for CS patients in the test cohort.
The DLRM can accurately predict the histologic grade and prognosis in CS.
计算机断层扫描（CT）和活检对于软骨肉瘤患者术前分级及预后评估可能并不充分。本研究旨在开发并验证一种基于CT的深度学习放射组学模型（DLRM），用于预测软骨肉瘤（CS）的组织学分级和预后。
纳入多中心211例CS患者（训练队列/测试队列，127/84）。开发了放射组学特征（RS）、深度学习特征（DLS）以及结合放射组学和深度学习特征的DLRM来预测分级。采用Kaplan-Meier生存分析评估模型预测分级与无复发生存期（RFS）的相关性。通过受试者操作特征曲线下面积（AUC）和Harrell一致性指数（C指数）评估模型性能。
在测试队列中预测分级时，DLRM（AUC，0.879；95%置信区间[CI]，0.802-0.956）优于（z = 2.773，P = 0.006）RS（AUC，0.715；95% CI，0.606-0.825）。根据DLRM分层的低级别和高级别患者之间，RFS显示出显著差异（对数秩检验，P<0.05）。在测试队列中预测CS患者的RFS时，DLRM的C指数（0.805；95% CI，0.694-0.916）高于RS（0.692，95% CI，0.540-0.844）。
DLRM能够准确预测CS的组织学分级和预后。
