基于瘤内瘤周超声组学特征预测三阴性乳腺癌

郑荣荣; 刘晓杰; 刘薇; 查海玲; 栗翠英

doi:10.19732/j.cnki.2096-6210.2026.02.007

您当前的位置：

首页 >

文章列表页 >

基于瘤内瘤周超声组学特征预测三阴性乳腺癌

论著 | 更新时间：2026-05-07

- 基于瘤内瘤周超声组学特征预测三阴性乳腺癌
- Prediction of triple-negative breast cancer via intratumoral and peritumoral ultrasound-based radiomics
- 肿瘤影像学 2026年35卷第2期页码：273-281
- 作者机构：
  
  1.南京医科大学第一附属医院超声诊断科，江苏南京 210029
  2.淮安市妇幼保健院超声科，江苏淮安 223002
- 作者简介：
  
  LI Cuiying E-mail: lynx_ko@163.com
- 基金信息：
  
  无
- DOI：10.19732/j.cnki.2096-6210.2026.02.007
  中图分类号： R737.9;R445.1
- 收稿：2025-11-21，
  
  修回：2026-02-10，
  
  纸质出版：2026-04-28
- 稿件说明：
移动端阅览
郑荣荣, 刘晓杰, 刘　薇, 等. 基于瘤内瘤周超声组学特征预测三阴性乳腺癌［J］. 肿瘤影像学, 2026, 35(2): 273-281.

ZHENG R R, LIU X J, LIU WCitation:, et al. Prediction of triple-negative breast cancer via intratumoral and peritumoral ultrasound-based radiomics［J］. Oncoradiology, 2026, 35(2): 273-281.
郑荣荣, 刘晓杰, 刘　薇, 等. 基于瘤内瘤周超声组学特征预测三阴性乳腺癌［J］. 肿瘤影像学, 2026, 35(2): 273-281. DOI： 10.19732/j.cnki.2096-6210.2026.02.007.

ZHENG R R, LIU X J, LIU WCitation:, et al. Prediction of triple-negative breast cancer via intratumoral and peritumoral ultrasound-based radiomics［J］. Oncoradiology, 2026, 35(2): 273-281. DOI： 10.19732/j.cnki.2096-6210.2026.02.007.

摘要

目的

运用超声组学技术提取三阴性乳腺癌（triple-negative breast cancer，TNBC）瘤内及瘤周4 mm区域的特征，结合多种机器学习分类器构建预测模型，以提高TNBC的诊断准确度。

方法

回顾并收集2017年4月—2023年7月南京医科大学第一附属医院经病理学检查确诊的乳腺癌患者的超声图像，将其分为训练集和测试集。对瘤内和瘤周4 mm环形区域进行勾画，提取形态学、纹理及一阶特征。采用最小冗余最大相关算法（minimal redundancy maximal relevance，mRMR）筛选特征后，基于随机森林（random forest）、极端随机树（extra-trees）、极端梯度提升（XGBoost）和轻量级梯度提升机（LightGBM）4类机器学习分类器，以及4种预测特征来源组合（包括仅来自瘤内区域的特征、仅来自瘤周区域的特征、瘤内和瘤周区域的图像融合特征以及瘤内区域特征与瘤周区域特征相结合的特征融合），构建16个超声组学模型，筛选最优预测模型，采用沙普利加性解释（SHapley Additive exPlanation，SHAP）揭示模型的重要特征。

结果

共纳入563例乳腺癌患者，其中TNBC 107例，非TNBC 456例。按7∶3随机分为训练集（394例）和测试集（169例）。分别从瘤内区域、瘤周区域以及瘤内与瘤周图像融合区域各提取1 561个超声组学特征。基于extra-trees的特征融合模型表现最佳，训练集、测试集的曲线下面积（area under curve，AUC）达0.952、0.857，而瘤内模型AUC为0.912、0.820，瘤周模型AUC为0.909、0.806，图像融合模型AUC为0.924、0.848，表明特征融合模型在预测TNBC方面优于单独瘤内或瘤周特征的模型。SHAP进一步揭示了瘤内区域的3个超声组学特征和瘤周区域的2个超声组学特征为模型中最具影响力的5个决定性因素。

结论

本研究强调了利用瘤内和瘤周的超声组学特征构建的extra-trees模型，在预测TNBC方面的潜力。

Abstract

Objective

To leverage radiomics based on ultrasonography to extract quantitative features from both the intratumoral volume and a 4 mm peritumoral rim of triple-negative breast cancer (TNBC)

and to integrate these features with multiple machine-learning classifiers to construct a high-accuracy diagnostic model.

Methods

A retrospective cohort of pathologically confirmed breast-cancer patients who underwent ultrasonography between April 2017 and July 2023 was collected and randomly split into training and testing sets. After manual segmentation of the intratumoral region and a 4 mm peritumoral annulus

radiomic descriptors (morphological

first-order

and texture features) were extracted. Following minimum redundancy maximum relevance (mRMR) feature selection

sixteen radiomics models were built by combining four machine-learning classifiers—random forest

extra-trees， XGBoost

and LightGBM—with four distinct feature-source strategies: intratumoral features only

peritumoral features only

fused images of both compartments

and concatenated intratumoral + peritumoral features. Model performance was evaluated by area under the receiver-operating-characteristic curve (AUC). SHapley Additive exPlanations (SHAP) were used to interpret the optimal model.

Results

A total of 563 breast cancer patients were in

cluded

consisting of 107 TNBC cases and 456 non-TNBC cases. Patients were randomly divided into a training set (

=394) and a test set (

=169) at a ratio of 7∶3. A total of 1 561 ultrasound radiomics features were extracted from each of the intratumoral

peritumoral

and image fusion regions

respectively. The extra-trees model trained on concatenated intratumoral and peritumoral features achieved the highest AUC (training set 0.952

test set 0.857)

outperforming intratumoral-only (0.912

0.820)

peritumoral-only (0.909

0.806)

and fused-image models (0.924

0.848). SHAP analysis identified three intratumoral and two peritumoral radiomic features as the five most influential determinants of TNBC prediction.

Conclusion

Integrating intratumoral and peritumoral ultrasound radiomics via an extra-trees classifier significantly enhances the non-invasive identification of TNBC

underscoring its translational potential in precision oncology.

关键词

Keywords

references

KIM J , HARPER A , MCCORMACK V , et al . Global patterns and trends in breast cancer incidence and mortality across 185 countries ［J］. Nat Med , 2025 , 31 ( 4 ): 1154 - 1162 .

CAI Y Z , DAI F Y , YE Y , et al . The global burden of breast cancer among women of reproductive age: a comprehensive analysis ［J］. Sci Rep , 2025 , 15 ( 1 ): 9347 .

KUMARI L , MISHRA L , PATEL P , et al . Emerging targeted therapeutic strategies for the treatment of triple-negative breast cancer ［J］. J Drug Target , 2023 , 31 ( 9 ): 889 - 907 .

VALLURI A R , CARTER G J , ROBRAHN I , et al . Triple-negative breast cancer: radiologic-pathologic correlation ［J］. J Breast Imaging , 2025 , 7 ( 3 ): 331 - 344 .

CHEN Z Y , ZHAO Y . The mechanism underlying metastasis in triple-negative breast cancer: focusing on the interplay between ferroptosis, epithelial-mesenchymal transition, and non-coding RNAs ［J］. Front Pharmacol , 2025 , 15 : 1437022 .

CHENG C , WANG Y , ZHAO J E , et al . Deep learning and radiomics in triple-negative breast cancer: predicting long-term prognosis and clinical outcomes ［J］. J Multidiscip Healthc , 2025 , 18 : 319 - 327 .

李晴晴 , 张良西 , 朱向明 , 等 . 基于术前超声影像组学及常规超声特征预测三阴性乳腺癌［J］. 赣南医学院学报 , 2022 ( 10 ): 1056 - 1061 .

LI Q Q , ZHANG L X , ZHU X M , et al . Prediction of triple-negative breast cancer based on preoperative ultrasonographic radiomics and conventional ultrasonic characteristics ［J］. J Gannan Med Univ , 2022 ( 10 ): 1056 - 1061 .

王　瑛 , 陈英格 , 叶素敏 , 等 . 超声影像组学标签预测三阴性乳腺癌的价值［J］. 中国医学影像学杂志 , 2021 , 29 ( 1 ): 35 - 38 .

WANG Y , CHEN Y G , YE S M , et al . Ultrasound-based radiomics in predicting triple-negative breast cancer ［J］. Chin J Med Imag , 2021 , 29 ( 1 ): 35 - 38 .

GUO Z J , ZHU Z Y , LIN X X , et al . Tumor microenvironment and immunotherapy for triple-negative breast cancer ［J］. Biomark Res , 2024 , 12 ( 1 ): 166 .

DING J , CHEN S L , SERRANO SOSA M , et al . Optimizing the peritumoral region size in radiomics analysis for sentinel lymph node status prediction in breast cancer ［J］. Acad Radiol , 2022 , 29 ( Suppl 1 ): S223 - S228 .

LIU W , LI L , DENG J , et al . A comprehensive approach for evaluating lymphovascular invasion in invasive breast cancer: leveraging multimodal MRI findings, radiomics, and deep learning analysis of intra- and peritumoral regions ［J］. Comput Med Imaging Graph , 2024 , 116 : 102415 .

YAO J J , JIA X H , ZHOU W , et al . Predicting axillary response to neoadjuvant chemotherapy using peritumoral and intratumoral ultrasound radiomics in breast cancer subtypes ［J］. iScience , 2024 , 27 ( 9 ): 110716 .

HUANG Z B , MO S J , WU H Y , et al . Optimizing breast cancer diagnosis with photoacoustic imaging: an analysis of intratumoral and peritumoral radiomics ［J］. Photoacoustics , 2024 , 38 : 100606 .

NORTON K A , JIN K , POPEL A S . Modeling triple-negative breast cancer heterogeneity: effects of stromal macrophages, fibroblasts and tumor vasculature ［J］. J Theor Biol , 2018 , 452 : 56 - 68 .

TZANAKAKIS G , GIATAGANA E M , KUSKOV A , et al . Proteoglycans in the pathogenesis of hormone-dependent cancers: mediators and effectors ［J］. Cancers , 2020 , 12 ( 9 ): 2401 .

DERAKHSHAN F , REIS-FILHO J S . Pathogenesis of triple-negative breast cancer ［J］. Annu Rev Pathol , 2022 , 17 : 181 - 204 .

YANG F , WANG Y C , LI Q , et al . Intratumor heterogeneity predicts metastasis of triple-negative breast cancer ［J］. Carcinogenesis , 2017 , 38 ( 9 ): 900 - 909 .

JIANG L , YOU C , XIAO Y , et al . Radiogenomic analysis reveals tumor heterogeneity of triple-negative breast cancer ［J］. Cell Rep Med , 2022 , 3 ( 7 ): 100694 .

LI J W , CAO Y C , ZHAO Z J , et al . Prediction for pathological and immunohistochemical characteristics of triple-negative invasive breast carcinomas: the performance comparison between quantitative and qualitative sonographic feature analysis ［J］. Eur Radiol , 2022 , 32 ( 3 ): 1590 - 1600 .

CHEN Q Y , XIA J G , ZHANG J . Identify the triple-negative and non-triple-negative breast cancer by using texture features of medicale ultrasonic image: a STROBE-compliant study ［J］. Medicine , 2021 , 100 ( 22 ): e25878 .

WU T , SULTAN L R , TIAN J W , et al . Machine learning for diagnostic ultrasound of triple-negative breast cancer ［J］. Breast Cancer Res Treat , 2019 , 173 ( 2 ): 365 - 373 .

MEHRAJ U , MUSHTAQ U , MIR M A , et al . Chemokines in triple-negative breast cancer heterogeneity: new challenges for clinical implications ［J］. Semin Cancer Biol , 2022 , 86 ( Pt 2 ): 769 - 783 .

XIE T W , GONG J , ZHAO Q F , et al . Development and validation of peritumoral vascular and intratumoral radiomics to predict pathologic complete responses to neoadjuvant chemotherapy in patients with triple-negative breast cancer ［J］. BMC Med Imaging , 2024 , 24 ( 1 ): 136 .

LUO S W , CHEN X B , YAO M X , et al . Intratumoral and peritumoral ultrasound-based radiomics for preoperative prediction of HER2-low breast cancer: a multicenter retrospective study ［J］. Insights Imaging , 2025 , 16 ( 1 ): 53 .

浏览量

下载量

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于多参数MRI生境成像的机器学习模型预测胃癌脉管浸润

靶向VEGFR2的小分子诊疗一体化探针体系⁶⁴Cu/¹⁷⁷Lu-DOTA-diZD在三阴性乳腺癌中的应用

超声人工智能在前列腺癌诊断中的应用进展

整合外周血免疫细胞与FS-T2WI影像组学优化肝细胞癌微血管侵犯预测模型研究

基于机器学习的影像组学模型预测局部晚期宫颈癌患者同步放化疗后结局