Scholarlink

Objectives <p>Noninvasive and accurate prediction of clinical staging is crucial for patients with cervical cancer (CC). The aim of our study was to develop a clinical model, a diffusion-weighted imaging (DWI)-based radiomics model, and a combined model for staging prediction.</p> Materials and methods <p>We retrospectively enrolled a total of 234 patients with histopathologically confirmed cervical cancer. The patients were divided into a training set (<i>n</i> = 163) and a testing set (<i>n</i> = 71). Radiomics features were extracted from tumor regions in DWI sequences. Clinical features were obtained through univariate and multivariate regression analyses. The clinical model and the DWI-based radiomics model were constructed using the selected features. A nomogram was also created to combine radiomics elements with clinical factors for predicting early-stage and advanced-stage CC.</p> Results <p>The area under the curve (AUC) values for the radiomics signature, which included features from the tumor’s region of interest (ROI) based on DWI, were 0.928 in the training set and 0.801 in the testing set. When the nomogram combined clinical data with the radiomics signature, the AUC values improved to 0.949 for the training set and 0.847 for the testing set in predicting staging.</p> Conclusions <p>The novel nomogram, which integrates clinical factors with radiomics features extracted from DWI sequences, demonstrates good performance in staging CC, assisting clinicians in making informed treatment decisions.</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

Clinical translation of a DWI-radiomics nomogram integrating serum biomarkers for pretreatment staging of cervical cancer

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月30日

查看全文
  • Nan Jiang,
  • Zhonghong Xin,
  • Yueyue Zhang,
  • Yuanqing Liu,
  • Meng Chen,
  • Xiaoxia Ping,
  • Qian Meng,
  • Chunhong Hu

摘要

Objectives

Noninvasive and accurate prediction of clinical staging is crucial for patients with cervical cancer (CC). The aim of our study was to develop a clinical model, a diffusion-weighted imaging (DWI)-based radiomics model, and a combined model for staging prediction.

Materials and methods

We retrospectively enrolled a total of 234 patients with histopathologically confirmed cervical cancer. The patients were divided into a training set (n = 163) and a testing set (n = 71). Radiomics features were extracted from tumor regions in DWI sequences. Clinical features were obtained through univariate and multivariate regression analyses. The clinical model and the DWI-based radiomics model were constructed using the selected features. A nomogram was also created to combine radiomics elements with clinical factors for predicting early-stage and advanced-stage CC.

Results

The area under the curve (AUC) values for the radiomics signature, which included features from the tumor’s region of interest (ROI) based on DWI, were 0.928 in the training set and 0.801 in the testing set. When the nomogram combined clinical data with the radiomics signature, the AUC values improved to 0.949 for the training set and 0.847 for the testing set in predicting staging.

Conclusions

The novel nomogram, which integrates clinical factors with radiomics features extracted from DWI sequences, demonstrates good performance in staging CC, assisting clinicians in making informed treatment decisions.