Scholarlink

<p>Our investigation focuses on developing and testing a radiomics nomogram based on contrast-enhanced mammography (CEM) and clinical factors to predict ductal carcinoma in situ (DCIS) in breast cancer.&#xa0;A retrospective analysis was performed on 731 breast cancer cases who underwent CEM examination and subsequent surgical treatment with complete pathological results, enrolled from five centers. Radiomics features were derived from both low-energy and recombined CEM images for each patient. The Minimum Redundancy Maximum Relevance (mRMR）and least absolute shrinkage and selection operator (LASSO) methods were used to select radiomics features. The radiomics signature (Rad-score) was calculated as a weighted linear combination of the most discriminative features. The univariate and multivariate logistic regression were used to select the clinical factors. A radiomics nomogram was established by integrating the Rad-score and independent clinical risk factors. The receiver operator characteristic curves (ROCs) and calibration curves were used to assess the performance of the radiomics nomogram.&#xa0;The Rad-score was calculated through the integration of 11 radiomics features. The radiomics nomogram was developed&#xa0;from Rad-score, age, menstrual status and background parenchymal enhancement (BPE) by logistic regression, which showed better predictive performance in both internal and pooled external test sets, with AUCs of 0.889 (95% confidence interval [CI]: 0.847-0.932) and 0.822 (95% CI: 0.630-1.000), respectively. The calibration curves exhibited excellent consistency between predicted and observed probabilities.&#xa0;The radiomics nomogram incorporated with CEM-based radiomics features, age, menstrual status and BPE showed acceptable performance in predicting the ductal carcinoma in situ in breast cancer. As a preliminary exploratory study, our findings require further validation in larger, multi-center external cohorts.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Contrast-Enhanced Mammography-Based Radiomics for Predicting Ductal Carcinoma In Situ in Breast Cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月29日

查看全文
  • Kun Zhang,
  • Juan Qiu,
  • Lijuan Wang,
  • Ning Mao,
  • Qin Wang,
  • Ping Sun,
  • Gang Chen,
  • Guangdong Qiao,
  • Simin Wang,
  • Kun Cao,
  • Yan Zhang,
  • Fan Lin,
  • Cong Xu

摘要

Our investigation focuses on developing and testing a radiomics nomogram based on contrast-enhanced mammography (CEM) and clinical factors to predict ductal carcinoma in situ (DCIS) in breast cancer. A retrospective analysis was performed on 731 breast cancer cases who underwent CEM examination and subsequent surgical treatment with complete pathological results, enrolled from five centers. Radiomics features were derived from both low-energy and recombined CEM images for each patient. The Minimum Redundancy Maximum Relevance (mRMR）and least absolute shrinkage and selection operator (LASSO) methods were used to select radiomics features. The radiomics signature (Rad-score) was calculated as a weighted linear combination of the most discriminative features. The univariate and multivariate logistic regression were used to select the clinical factors. A radiomics nomogram was established by integrating the Rad-score and independent clinical risk factors. The receiver operator characteristic curves (ROCs) and calibration curves were used to assess the performance of the radiomics nomogram. The Rad-score was calculated through the integration of 11 radiomics features. The radiomics nomogram was developed from Rad-score, age, menstrual status and background parenchymal enhancement (BPE) by logistic regression, which showed better predictive performance in both internal and pooled external test sets, with AUCs of 0.889 (95% confidence interval [CI]: 0.847-0.932) and 0.822 (95% CI: 0.630-1.000), respectively. The calibration curves exhibited excellent consistency between predicted and observed probabilities. The radiomics nomogram incorporated with CEM-based radiomics features, age, menstrual status and BPE showed acceptable performance in predicting the ductal carcinoma in situ in breast cancer. As a preliminary exploratory study, our findings require further validation in larger, multi-center external cohorts.