Scholarlink

Purpose <p>To evaluate a combined model incorporating T2-weighted imaging (T2WI)-based radiomics signature and apparent diffusion coefficient (ADC) histogram features for predicting kirsten rat sarcoma virus oncogene (KRAS) mutation status in rectal cancer patients.</p> Methods <p>220 patients with pathologically confirmed rectal adenocarcinoma from Center I (training set: <i>n</i> = 154; internal test set: <i>n</i> = 66) and 66 from Center II (external test set) were retrospectively included. A total of 851 radiomic features from T2WI and 20 ADC histogram features from diffusion-weighted imaging (DWI) were extracted. These two sets of features underwent separate feature selection and were then combined to construct a classification model for KRAS prediction. Model performance was evaluated using ROC curve analysis based on 1000 repeated train–test splits, and differences in AUCs were assessed using the Wilcoxon signed-rank test to account for paired performance distributions. <i>P</i> &lt; 0.05 was considered statistically significant.</p> Results <p>The combined model achieved the highest performance in the internal test set, with a median AUC of 0.823 [IQR: 0.797–0.848], outperforming the radiomics-only model (AUC = 0.754 [0.720–0.788]) and the ADC-only model (AUC = 0.607 [0.561–0.696]) (<i>P</i> &lt; 0.001). In the external test set, it maintained superior performance (AUC = 0.772 [0.743–0.799]) and significantly outperformed the radiomics-only (AUC = 0.718 [0.682–0.749]) and ADC-only models (AUC = 0.632 [0.616–0.694]) (<i>P</i> &lt; 0.001).</p> Conclusion <p>The combined model demonstrated robust performance for predicting KRAS mutation status in rectal cancer and holds promise as a noninvasive adjunct to genetic testing in clinical settings.</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

Preoperative prediction of KRAS mutation in rectal cancer using a combined T2-weighted imaging radiomics and volumetric apparent diffusion coefficient histogram model

  • 期刊论文

  • 发表日期：

    2026年2月18日

查看全文
  • Qiaoyu Liu,
  • Rencheng Zheng,
  • Zhangwei Yang,
  • Jingqi Zhu

摘要

Purpose

To evaluate a combined model incorporating T2-weighted imaging (T2WI)-based radiomics signature and apparent diffusion coefficient (ADC) histogram features for predicting kirsten rat sarcoma virus oncogene (KRAS) mutation status in rectal cancer patients.

Methods

220 patients with pathologically confirmed rectal adenocarcinoma from Center I (training set: n = 154; internal test set: n = 66) and 66 from Center II (external test set) were retrospectively included. A total of 851 radiomic features from T2WI and 20 ADC histogram features from diffusion-weighted imaging (DWI) were extracted. These two sets of features underwent separate feature selection and were then combined to construct a classification model for KRAS prediction. Model performance was evaluated using ROC curve analysis based on 1000 repeated train–test splits, and differences in AUCs were assessed using the Wilcoxon signed-rank test to account for paired performance distributions. P < 0.05 was considered statistically significant.

Results

The combined model achieved the highest performance in the internal test set, with a median AUC of 0.823 [IQR: 0.797–0.848], outperforming the radiomics-only model (AUC = 0.754 [0.720–0.788]) and the ADC-only model (AUC = 0.607 [0.561–0.696]) (P < 0.001). In the external test set, it maintained superior performance (AUC = 0.772 [0.743–0.799]) and significantly outperformed the radiomics-only (AUC = 0.718 [0.682–0.749]) and ADC-only models (AUC = 0.632 [0.616–0.694]) (P < 0.001).

Conclusion

The combined model demonstrated robust performance for predicting KRAS mutation status in rectal cancer and holds promise as a noninvasive adjunct to genetic testing in clinical settings.