基于瘤内与瘤周DBT影像组学列线图预测三阴性乳腺癌的价值  

The Value of Predicting Triple-Negative Breast Cancer Based on Intratumoural and Peritumoural DBT Radiomics Nomogram

作　　者：陈修婷 李杰 曹明娜 朱芸 高之振 CHEN Xiuting;LI Jie;CAO Mingna(Department of Radiology,the First Affiliated Hospital of Bengbu Medical University,Bengbu,Anhui Province 233004,P.R.China)

机构地区：[1]蚌埠医科大学第一附属医院放射科,233004 [2]蚌埠医科大学研究生院,233030

出　　处：《临床放射学杂志》2025年第3期435-443,共9页Journal of Clinical Radiology

基　　金：安徽省教育厅自然科学重点项目(编号:2022AH051473);蚌埠医科大学自然科学重点项目(编号:2021byzd027)基金资助项目。

摘　　要：目的探讨基于数字乳腺断层合成X线摄影(DBT)的瘤内联合瘤周影像组学特征及临床因素构建的列线图预测肿块型三阴性乳腺癌(TNBC)的价值。方法回顾性搜集符合纳排标准的163例乳腺癌患者的DBT影像学信息及临床病理指标,以7∶3的比例随机分为训练集(n=114)和验证集(n=49)。医师手动勾画瘤内感兴趣区域(ROI),外扩3 mm获取瘤周ROI;提取并筛选影像组学特征,利用逻辑回归(LR)分类器构建瘤内、瘤周、瘤内+瘤周影像组学模型;以效能最高的影像组学模型的Pred-score联合临床独立预测因子构建列线图模型。Delong检验用于比较各模型受试者工作特征曲线曲线下面积(AUC)差异。受试者工作特征(ROC)曲线、决策曲线及校准曲线评估模型性能。结果肿块触诊活动度(P<0.001)和肿块形态(P=0.012)是TNBC的独立预测因子;瘤内+瘤周模型较单一瘤内、瘤周模型诊断性能更优;列线图模型实现了最佳预测效能,其训练集AUC、灵敏度、特异度、准确度分别为0.907、82.1%、85.3%、85.6%,验证集分别为0.856、81.8%、81.6%、82.9%。结论基于DBT瘤内联合瘤周影像组学特征及临床指标的列线图能于术前有效鉴别TNBC,可作为一种无创预测方法指导临床决策。Objective To investigate the value of the nomogram based on intratumoral and peritumoral radiomics features and clinical factors of digital breast tomosynthesis(DBT)in predicting mass-like triple-negative breast cancer(TNBC).Methods DBT imaging data and clinical-pathological parameters were retrospectively collected from 163 breast cancer patients who met the inclusion and exclusion criteria.They were randomly divided into a training set(n=114)and a validation set(n=49)in a 7∶3 ratio.The intratumoral region of interest(ROI)was manually delineated by a physician,and the peritumoral ROI was obtained by expanding 3 mm outward from the tumor boundary.Radiomics features were extracted and screened,and logistic regression(LR)classifiers were used to construct intratumoral,peritumoral,and combined intratumoral+peritumoral radiomics models.The Pred-score of the radiomics model with the highest predictive efficacy was selected and combined with clinically independent predictors to develop a nomogram model.The DeLong test was used to compare differences in AUC across models.ROC curves,decision curves,and calibration curves were used to evaluate model performance.Results Tumor palpation mobility(P<0.001)and tumor morphology(P=0.012)were identified as independent predictors of TNBC.The combined intratumoral+peritumoral model demonstrated superior diagnostic performance compared to either the intratumoral or peritumoral model alone.The nomogram model achieved the best predictive performance,with an AUC of 0.907,sensitivity of 82.1%,specificity of 85.3%,and accuracy of 85.6%in the training set,and an AUC of 0.856,sensitivity of 81.8%,specificity of 81.6%,and accuracy of 82.9%in the validation set.Conclusion The nomogram based on DBT intratumoral and peritumoral radiomics features and clinical indicators can effectively distinguish TNBC before surgery and can be used as a non-invasive prediction method to guide clinical decision-making.

关 键 词：三阴性乳腺癌 瘤周 数字乳腺断层合成X线摄影 影像组学 列线图 

分 类 号：R730.44[医药卫生—肿瘤] R737.9[医药卫生—临床医学]