基于深度学习的多组学模型用于结直肠癌术后远处转移风险分层
这项回顾性研究纳入两个中心521例根治性手术的结直肠癌患者,利用术前CT和术后HE染色切片构建深度学习影像组学与病理组学模型及列线图。训练集中Nomogram 1和Nomogram 2的AUC分别为0.938和0.930,显著优于传统影像和病理模型,Kaplan-Meier分析证实列线图可有效分层3年无病生存。
要点
- 基于ResNet-101提取深度特征构建的DLRS和DLPS模型在训练集AUC分别达0.931和0.914。
- Nomogram 1(影像+DLRS)和Nomogram 2(病理+DLPS)训练集AUC分别为0.938和0.930。
- 列线图模型可对结直肠癌术后患者进行有效的3年无病生存风险分层。
核心结论
- 基于深度学习的多组学模型对结直肠癌术后远处转移具有较高预测准确性,列线图模型能可靠地进行无病生存风险分层。
研究设计与样本
- 回顾性研究,纳入两个中心共521例接受根治性手术的结直肠癌患者,收集术前CT和术后HE染色切片;中心1的381例按7:3分为训练集和内部验证集,中心2的140例作为独立外部验证集;采用ResNet-101骨干网络提取深度特征。
诊断标准与关键数值
- CT报告的T(cT)分期(OR=2.00,P=0.006)和CT报告的N(cN)分期(OR=1.63,P=0.023)为构建影像学模型的独立影像预测因子。
- pN分期(OR=1.91,P=0.003)和神经周围侵犯(OR=2.07,P=0.030)为构建病理学模型的病理预测因子。
- DLRS和DLPS分别纳入28个和30个深度特征。
- 训练集中影像学、病理学、DLRS、DLPS、Nomogram 1和Nomogram 2模型的AUC分别为0.657、0.687、0.931、0.914、0.938和0.930。
- DeLong检验显示DLRS、DLPS及两个列线图模型均显著优于传统模型(P<.05)。
临床应用与阅片要点
- 作者认为基于深度学习的多组学模型对结直肠癌术后远处转移预测具有较高准确性。
- 作者认为列线图模型能够对结直肠癌患者进行可靠的无病生存风险分层。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: To develop deep learning-based multiomics models for predicting postoperative distant metastasis (DM) and evaluating survival prognosis in colorectal cancer (CRC) patients.
MATERIALS AND METHODS: This retrospective study included 521 CRC patients who underwent curative surgery at two centers. Preoperative CT and postoperative hematoxylin-eosin (HE) stained slides were collected. A total of 381 patients from Center 1 were split (7:3) into training and internal validation sets; 140 patients from Center 2 formed the independent external validation set. Patients were grouped based on DM status during follow-up. Radiological and pathological models were constructed using independent imaging and pathological predictors. Deep features were extracted with a ResNet-101 backbone to build deep learning radiomics (DLRS) and deep learning pathomics (DLPS) models. Two integrated models were developed: Nomogram 1 (radiological + DLRS) and Nomogram 2 (pathological + DLPS).
RESULTS: CT- reported T (cT) stage (OR=2.00, P=0.006) and CT-reported N (cN) stage (OR=1.63, P=0.023) were identified as independent radiologic predictors for building the radiological model; pN stage (OR=1.91, P=0.003) and perineural invasion (OR=2.07, P=0.030) were identified as pathological predictors for building the pathological model. DLRS and DLPS incorporated 28 and 30 deep features, respectively. In the training set, area under the curve (AUC) for radiological, pathological, DLRS, DLPS, Nomogram 1, and Nomogram 2 models were 0.657, 0.687, 0.931, 0.914, 0.938, and 0.930. DeLong's test showed DLRS, DLPS, and both nomograms significantly outperformed conventional models (P<.05). Kaplan-Meier analysis confirmed effective 3-year disease-free survival (DFS) stratification by the nomograms.
CONCLUSION: Deep learning-based multiomics models provided high accuracy for postoperative DM prediction. Nomogram models enabled reliable DFS risk stratification in CRC patients.
引用
- Yao X, Han X, Huang D, et al. Deep Learning Based Multiomics Model for Risk Stratification of Postoperative Distant Metastasis in Colorectal Cancer. Acad Radiol. 2025. PMID: 40912950. doi:10.1016/j.acra.2025.08.040
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。