基于超声的临床-影像组学列线图预测局灶性睾丸肿瘤组织学亚型的研究
这项回顾性研究纳入148例经病理证实的睾丸肿瘤患者(120例睾丸生殖细胞肿瘤,其中65例精原细胞瘤、55例非精原细胞瘤),构建基于超声的临床-影像组学列线图,用于术前鉴别睾丸生殖细胞肿瘤与非生殖细胞肿瘤并区分精原细胞瘤与非精原细胞瘤。生殖细胞肿瘤列线图在训练集和验证集AUC均为0.89;精原细胞瘤列线图训练集AUC为0.93、验证集AUC为0.91,优于Radscore模型和临床模型。
要点
- 超声临床-影像组学列线图有潜力无创鉴别睾丸肿瘤组织学亚型。
- 精原细胞瘤列线图训练集AUC 0.93、验证集AUC 0.91,优于Radscore和临床模型。
- 生殖细胞肿瘤列线图训练集与验证集AUC均为0.89。
核心结论
- 基于超声的临床-影像组学列线图有潜力无创区分睾丸肿瘤的组织学亚型。
研究设计与样本
- 回顾性研究纳入148例经病理证实的睾丸肿瘤患者,其中睾丸生殖细胞肿瘤120例(精原细胞瘤65例、非精原细胞瘤55例),所有患者术前行超声检查,收集临床信息、超声特征和影像组学特征,经特征选择构建Radscore模型,采用单因素和多因素logistic回归分析确定独立危险因素并构建列线图,用ROC曲线和决策曲线分析评估模型。
诊断标准与关键数值
- 睾丸生殖细胞肿瘤影像组学列线图模型在训练集和验证集的AUC均为0.89。
- 精原细胞瘤影像组学列线图模型训练集AUC为0.93、验证集AUC为0.91,优于Radscore模型和临床模型。
- 校准曲线显示列线图估计值与实际观察值一致,决策曲线分析验证了联合模型的临床价值。
临床应用与阅片要点
- 作者认为基于超声的临床-影像组学列线图有潜力无创鉴别睾丸肿瘤组织学亚型。
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
OBJECTIVES: Testicular tumors are the most common solid malignancy among males aged 15-35. This study aimed to establish an ultrasound (US) based clinical-radiomic nomogram for the preoperative prediction of testicular tumors histologic subtypes, differentiating testicular germ cell tumors (TGCTs) from testicular non-germ cell tumors (TNGCTs) and then differentiating seminomas (SGCTs) from non-seminomatous tumors (NSGCTs).
METHODS: This retrospective study included 148 patients with testicular tumors confirmed by pathology, with 120 cases of TGCTs, including 65 SGCTs and 55 NSGCTs. All patients underwent preoperative ultrasound examinations, and data on clinical information, US features, and radiomics features were collected. The Radscore model was constructed after feature selection. Independent risk factors were identified using univariate and multivariate logistic regression analysis. The nomogram model was assessed using the receiver operating characteristic (ROC) curve analysis and decision curve analysis (DCA).
RESULTS: The TGCTs radiomics nomogram model achieved AUCs of 0.89 in both the training and validation datasets. The SGCTs radiomics nomogram model achieved AUCs of 0.93 in the training dataset and 0.91 in the validation dataset, surpassing the predictive performance of both Radscore and clinical models. The calibration curves showed that the nomogram estimation was consistent with the actual observations. DCA also verified the clinical value of the combined model.
CONCLUSIONS: The ultrasound-based clinical-radiomics nomogram has the potential to non-invasively discriminate the histologic subtypes of testicular tumors.
引用
- Lin T, Chen S, Miao S, et al. Development of an Ultrasound-Based Clinical-Radiomic Nomogram for Predicting Histologic Subtypes in Focal Testicular Tumors. J Ultrasound Med. 2025. PMID: 40177912. doi:10.1002/jum.16701
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。