Scholarlink

Objective <p>The exploration of clinical characteristics associated with osteoporosis (OP) is crucial due to its high incidence and complex pathogenesis. Therefore, this study analyzed clinical characteristics for OP in a large cohort of middle-aged and elderly patients in Kunming and constructed a nomogram model through OP-related characteristics.</p> Methods <p>Following inclusion and exclusion, we obtained 1,220 middle-aged and elderly patients (480 OP, 740 non-OP) from 1,847 patients and analyzed approximately 200 clinical characteristics for significant differences. We randomly split the 1,220 patients into a 7:3 training and validation set. In the training set, we used univariate and multivariate logistic regression, along with Least absolute shrinkage and selection operator (LASSO) regression, to identify OP-related clinical characteristics, which were then used to construct a nomogram diagnostic model. The nomogram model’s performance was validated via ROC curves, calibration curves and DCA curves.</p> Results <p>We identified over 50 clinical characteristics that showed significant differences between the OP and non-OP groups. Using machine learning, we screened for 18 characteristics closely associated with OP, including <i>Age</i>, <i>Sex</i>, <i>Smoke</i>, <i>Left hand muscle strength</i>, <i>Right hand muscle strength</i>, <i>Height Shorter</i>, <i>Creatinine</i>, and <i>Bone mineral density-</i>related characteristics, among others. The nomogram model, based on 18 clinical characteristics, achieved AUCs of 0.998 (training set) and 0.994 (validation set) for ROC curves, demonstrating excellent diagnostic performance supported by calibration and DCA curves.</p> Conclusion <p>This study identified 18 OP-related clinical characteristics and constructed a nomogram model based on these characteristics, which exhibited excellent discriminatory power between OP and non-OP patients.</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

Development and validation of a nomogram for osteoporosis based on clinical characteristics in Kunming

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月11日

查看全文
  • Xiaohan Tan,
  • Chai Yuan,
  • Jiabao Liao,
  • Jie Zhao,
  • Yuanliang Ai,
  • Hanyu He,
  • Weibo Wen,
  • Xuehua Xie

摘要

Objective

The exploration of clinical characteristics associated with osteoporosis (OP) is crucial due to its high incidence and complex pathogenesis. Therefore, this study analyzed clinical characteristics for OP in a large cohort of middle-aged and elderly patients in Kunming and constructed a nomogram model through OP-related characteristics.

Methods

Following inclusion and exclusion, we obtained 1,220 middle-aged and elderly patients (480 OP, 740 non-OP) from 1,847 patients and analyzed approximately 200 clinical characteristics for significant differences. We randomly split the 1,220 patients into a 7:3 training and validation set. In the training set, we used univariate and multivariate logistic regression, along with Least absolute shrinkage and selection operator (LASSO) regression, to identify OP-related clinical characteristics, which were then used to construct a nomogram diagnostic model. The nomogram model’s performance was validated via ROC curves, calibration curves and DCA curves.

Results

We identified over 50 clinical characteristics that showed significant differences between the OP and non-OP groups. Using machine learning, we screened for 18 characteristics closely associated with OP, including Age, Sex, Smoke, Left hand muscle strength, Right hand muscle strength, Height Shorter, Creatinine, and Bone mineral density-related characteristics, among others. The nomogram model, based on 18 clinical characteristics, achieved AUCs of 0.998 (training set) and 0.994 (validation set) for ROC curves, demonstrating excellent diagnostic performance supported by calibration and DCA curves.

Conclusion

This study identified 18 OP-related clinical characteristics and constructed a nomogram model based on these characteristics, which exhibited excellent discriminatory power between OP and non-OP patients.