Scholarlink

Background <p>The ultrasound diagnostic system for extrathyroidal extension (ETE) of papillary thyroid carcinoma (PTC) has not been thoroughly explored. To develop and validate a nomogram model based on ultrasound features to predict ETE of papillary thyroid carcinoma for preoperative assessment.</p> Methods <p>The training set retrospectively included 560 patients from two hospitals with preoperative ultrasound images showing capsule contact and confirmed as unifocal PTC by surgical pathology. The external validation set prospectively included 150 PTC patients with similar features and dynamic ultrasound videos. Univariate and multivariate logistic regression analyses were used to identify independent predictors of ETE in PTC, and an ETE nomogram prediction model was constructed to predict the risk of ETE in capsule-contacting PTC. The predictive efficiency of the model was evaluated using receiver operating characteristic (ROC) curve and calibration curves, and the clinical value of the model was determined through decision curve analysis (DCA).</p> Results <p>Among 710 capsule-contacting unifocal PTC patients, the incidence of ETE was 66.62% (473/710). Independent predictors of ETE were: Capsule bulging (OR = 8.951, 95%CI: 5.192–15.134), capsule contact angle ≥ 90° (OR = 2.331, 95%CI: 1.405–3.868), capsule contact extent ≥ 25% (OR = 5.708, 95%CI: 3.429–9.503), irregular morphology (OR = 1.856, 95%CI: 1.114–3.094), and coarse margins (OR = 4.198, 95%CI: 2.396–7.352). Based on these factors, an ETE nomogram diagnostic prediction model for PTC was established. The model’s ROC curve demonstrated an area under the curve (AUC) of 0.887 (95% CI: 0.857–0.917), with diagnostic sensitivity, specificity, and accuracy of 0.811, 0.799 and 0.807, respectively. The AUC of the external validation set was 0.896 (95% CI: 0.847–0.945), with diagnostic sensitivity, specificity, and accuracy of 0.862, 0.762, and 0.820, respectively. The calibration curve showed good consistency between the predicted and actual probabilities of ETE. DCA showed that the model had good clinical application value.</p> Conclusion <p>The ETE nomogram scoring prediction model based on conventional ultrasound features can provide a relatively convenient and intuitive preoperative quantitative assessment of ETE in PTC, serving as a reference for clinical decision-making.</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

Ultrasound feature-based nomogram model for predicting extrathyroidal extension in papillary thyroid carcinoma

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年7月18日

查看全文
  • Dong Guo,
  • Chen Chen,
  • Yin Zheng,
  • Yue Shan,
  • Shifei Huang,
  • Tianhan Zhou,
  • Yefei Yao,
  • Zhengxian Zhang,
  • Lu Wang,
  • Dong Xu

摘要

Background

The ultrasound diagnostic system for extrathyroidal extension (ETE) of papillary thyroid carcinoma (PTC) has not been thoroughly explored. To develop and validate a nomogram model based on ultrasound features to predict ETE of papillary thyroid carcinoma for preoperative assessment.

Methods

The training set retrospectively included 560 patients from two hospitals with preoperative ultrasound images showing capsule contact and confirmed as unifocal PTC by surgical pathology. The external validation set prospectively included 150 PTC patients with similar features and dynamic ultrasound videos. Univariate and multivariate logistic regression analyses were used to identify independent predictors of ETE in PTC, and an ETE nomogram prediction model was constructed to predict the risk of ETE in capsule-contacting PTC. The predictive efficiency of the model was evaluated using receiver operating characteristic (ROC) curve and calibration curves, and the clinical value of the model was determined through decision curve analysis (DCA).

Results

Among 710 capsule-contacting unifocal PTC patients, the incidence of ETE was 66.62% (473/710). Independent predictors of ETE were: Capsule bulging (OR = 8.951, 95%CI: 5.192–15.134), capsule contact angle ≥ 90° (OR = 2.331, 95%CI: 1.405–3.868), capsule contact extent ≥ 25% (OR = 5.708, 95%CI: 3.429–9.503), irregular morphology (OR = 1.856, 95%CI: 1.114–3.094), and coarse margins (OR = 4.198, 95%CI: 2.396–7.352). Based on these factors, an ETE nomogram diagnostic prediction model for PTC was established. The model’s ROC curve demonstrated an area under the curve (AUC) of 0.887 (95% CI: 0.857–0.917), with diagnostic sensitivity, specificity, and accuracy of 0.811, 0.799 and 0.807, respectively. The AUC of the external validation set was 0.896 (95% CI: 0.847–0.945), with diagnostic sensitivity, specificity, and accuracy of 0.862, 0.762, and 0.820, respectively. The calibration curve showed good consistency between the predicted and actual probabilities of ETE. DCA showed that the model had good clinical application value.

Conclusion

The ETE nomogram scoring prediction model based on conventional ultrasound features can provide a relatively convenient and intuitive preoperative quantitative assessment of ETE in PTC, serving as a reference for clinical decision-making.