Scholarlink

Purpose <p>This study aimed to develop a predictive model for diminished ovarian reserve (DOR) in reproductive-aged women with Hashimoto’s thyroiditis (HT) by integrating transvaginal three-dimensional ultrasound and clinical multimodal data, to enable early identification and accurate risk assessment of DOR, and provide a visualized decision-support tool for personalized reproductive health management.</p> Methods <p>A prospective cross-sectional study enrolled 300 eligible reproductive-aged women with HT, who underwent transvaginal three-dimensional ultrasound to measure bilateral ovarian volume, blood flow parameters, and antral follicle count (AFC). Participants were randomized at a 7:3 ratio into training (<i>n</i> = 210) and validation (<i>n</i> = 90) sets. LASSO regression was used to identify independent DOR risk factors for nomogram construction, with performance assessed using ROC curves, calibration curves, and decision curve analysis (DCA).</p> Results <p>Four independent risk factors were identified: pelvic inflammation history, ovarian vascularization index (VI), bilateral AFC, and positive thyroid globulin antibody (TGAb). The nomogram exhibited a significantly higher AUC than each individual factor (Delong’s test, all <i>p</i> &lt; 0.05), with AUC values of 0.989 (training set) and 0.985 (validation set), corresponding sensitivities of 0.964 and 0.882, and specificities of 0.930 and 0.974. Calibration curves indicated strong concordance between predicted and actual DOR probabilities, and DCA showed greater net benefit across threshold probabilities.</p> Conclusion <p>The nomogram integrating transvaginal three-dimensional ultrasound parameters and clinical indicators demonstrates excellent predictive accuracy for assessing DOR risk in reproductive-aged women with HT, providing a practical, visualized tool for personalized risk evaluation and reproductive health management.</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

Multi-dimensional predictive model for diminished ovarian reserve in Hashimoto’s thyroiditis: development and application

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月30日

查看全文
  • Jiemin Chen,
  • Huohu Zhong,
  • Yanyan Huang,
  • Kailing Tan,
  • Guorong Lyu,
  • Qiuyue Chen

摘要

Purpose

This study aimed to develop a predictive model for diminished ovarian reserve (DOR) in reproductive-aged women with Hashimoto’s thyroiditis (HT) by integrating transvaginal three-dimensional ultrasound and clinical multimodal data, to enable early identification and accurate risk assessment of DOR, and provide a visualized decision-support tool for personalized reproductive health management.

Methods

A prospective cross-sectional study enrolled 300 eligible reproductive-aged women with HT, who underwent transvaginal three-dimensional ultrasound to measure bilateral ovarian volume, blood flow parameters, and antral follicle count (AFC). Participants were randomized at a 7:3 ratio into training (n = 210) and validation (n = 90) sets. LASSO regression was used to identify independent DOR risk factors for nomogram construction, with performance assessed using ROC curves, calibration curves, and decision curve analysis (DCA).

Results

Four independent risk factors were identified: pelvic inflammation history, ovarian vascularization index (VI), bilateral AFC, and positive thyroid globulin antibody (TGAb). The nomogram exhibited a significantly higher AUC than each individual factor (Delong’s test, all p < 0.05), with AUC values of 0.989 (training set) and 0.985 (validation set), corresponding sensitivities of 0.964 and 0.882, and specificities of 0.930 and 0.974. Calibration curves indicated strong concordance between predicted and actual DOR probabilities, and DCA showed greater net benefit across threshold probabilities.

Conclusion

The nomogram integrating transvaginal three-dimensional ultrasound parameters and clinical indicators demonstrates excellent predictive accuracy for assessing DOR risk in reproductive-aged women with HT, providing a practical, visualized tool for personalized risk evaluation and reproductive health management.