Scholarlink

<p>This study aims to establish and validate a multi-dimensional clinical indicator-based risk prediction model for prostate cancer biopsy decision-making, and to develop an associated online visualization tool. This retrospective study enrolled 718 prostate biopsy patients from three Grade-A tertiary hospitals in Wuhan, with biopsy procedures spanning from October 2021 through October 2024. The data were randomly divided into training set and validation set at a ratio of 7∶3. The training set was divided into two groups based on the presence or absence of prostate cancer. LASSO regression was used for variable selection, Logistic regression analysis was used to identify the risk factors of prostate cancer, and a nomogram prediction model was constructed based on the results. Then the receiver operating characteristic (ROC) curve was used to evaluate the discrimination of the model. The calibration curve and Hosmer-Lemeshow test were used to verify the calibration of the model. Finally, the decision curve analysis (DCA) was used to evaluate the clinical application value of the model. In addition, an online visualization tool was generated using the Shiny package from R Studio. Prostate cancer was diagnosed in 280 patients, accounting for 38.9% of the study cohort. The logistic regression analysis identified age &gt; 65 years, prostate volume &lt; 35&#xa0;ml, PI-RADS scores &gt; 3, free prostate specific antigen / total prostate specific antigen &lt; 0.16, prostate-specific antigen density &gt; 0.15, and platelet &lt; 125 mmol/L as significant risk factors for prostate cancer (<i>P</i> &lt; 0.05). The discrimination efficiency of the prediction model was verified by ROC curve, and the area under the ROC curve was 0.763 (95%<i>CI</i>, 0.721–0.805). In terms of calibration, the calibration curve showed that the predicted value was in good agreement with the observed value, which was confirmed by Hosmer-Lemeshow test results (χ<sup>2</sup> = 11.553, <i>P</i> = 0.116). DCA further verified the clinical application value of the prediction model. Development of online visual tools can promote the clinical application of forecasting model (see the link: https://xiafuhai.shinyapps.io/dynnomapp/). This study successfully developed and validated the first multidimensional risk prediction model for prostate cancer biopsy decision-making specifically tailored for the Chinese population. The model demonstrated improved diagnostic efficacy compared with single clinical indicators. Through clinical implementation via nomogram visualization and an online interactive tool, the model exhibited moderate-to-good discrimination and calibration performance. This model provides a quantitative risk assessment tool for patients with clinically suspected prostate cancer who meet the indications for biopsy, and facilitates more accurate and targeted decision-making regarding prostate needle biopsy for both clinicians and 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

Construction, validation, and visualization tool development of a risk prediction model for prostate cancer biopsy decision-making

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年7月31日

查看全文
  • Qiang Li,
  • Fuhai Xia,
  • Xi Chen,
  • Li Li,
  • Xiaoshuang He,
  • Yingwu Zhang,
  • Chuanqi Liu,
  • Xuan Mao,
  • Baoguo Li,
  • Rui Chen,
  • Jing Li,
  • Yonglian Guo,
  • Adan Fu

摘要

This study aims to establish and validate a multi-dimensional clinical indicator-based risk prediction model for prostate cancer biopsy decision-making, and to develop an associated online visualization tool. This retrospective study enrolled 718 prostate biopsy patients from three Grade-A tertiary hospitals in Wuhan, with biopsy procedures spanning from October 2021 through October 2024. The data were randomly divided into training set and validation set at a ratio of 7∶3. The training set was divided into two groups based on the presence or absence of prostate cancer. LASSO regression was used for variable selection, Logistic regression analysis was used to identify the risk factors of prostate cancer, and a nomogram prediction model was constructed based on the results. Then the receiver operating characteristic (ROC) curve was used to evaluate the discrimination of the model. The calibration curve and Hosmer-Lemeshow test were used to verify the calibration of the model. Finally, the decision curve analysis (DCA) was used to evaluate the clinical application value of the model. In addition, an online visualization tool was generated using the Shiny package from R Studio. Prostate cancer was diagnosed in 280 patients, accounting for 38.9% of the study cohort. The logistic regression analysis identified age > 65 years, prostate volume < 35 ml, PI-RADS scores > 3, free prostate specific antigen / total prostate specific antigen < 0.16, prostate-specific antigen density > 0.15, and platelet < 125 mmol/L as significant risk factors for prostate cancer (P < 0.05). The discrimination efficiency of the prediction model was verified by ROC curve, and the area under the ROC curve was 0.763 (95%CI, 0.721–0.805). In terms of calibration, the calibration curve showed that the predicted value was in good agreement with the observed value, which was confirmed by Hosmer-Lemeshow test results (χ2 = 11.553, P = 0.116). DCA further verified the clinical application value of the prediction model. Development of online visual tools can promote the clinical application of forecasting model (see the link: https://xiafuhai.shinyapps.io/dynnomapp/). This study successfully developed and validated the first multidimensional risk prediction model for prostate cancer biopsy decision-making specifically tailored for the Chinese population. The model demonstrated improved diagnostic efficacy compared with single clinical indicators. Through clinical implementation via nomogram visualization and an online interactive tool, the model exhibited moderate-to-good discrimination and calibration performance. This model provides a quantitative risk assessment tool for patients with clinically suspected prostate cancer who meet the indications for biopsy, and facilitates more accurate and targeted decision-making regarding prostate needle biopsy for both clinicians and patients.