文献速递前列腺癌

基于机器学习的MRI成像用于前列腺癌诊断:系统综述与荟萃分析

Machine learning-based MRI imaging for prostate cancer diagnosis: systematic review and meta-analysis.
PubMed 文献 · Prostate Cancer Prostatic Dis · 2025-07-28

该系统综述与荟萃分析纳入12项研究、3474例患者,评估基于机器学习的MRI成像对前列腺癌良恶性鉴别及临床显著前列腺癌(csPCa,Gleason评分≥7)的诊断价值。结果显示,良恶性鉴别合并敏感度0.92、特异度0.90、AUC 0.96;csPCa诊断合并敏感度0.83、特异度0.73、AUC 0.86。结论认为诊断准确性良好,仍需深入研究验证。

要点

  • 机器学习MRI鉴别前列腺癌良恶性合并敏感度0.92、特异度0.90、AUC 0.96。
  • 检测csPCa合并敏感度0.83、特异度0.73、AUC 0.86。
  • 结论认为诊断准确性良好,但需进一步研究验证。

核心结论

  • 基于机器学习的MRI成像对前列腺癌良恶性鉴别和csPCa诊断均显示出良好的诊断准确性。
  • 作者指出仍需进一步深入研究以验证上述发现。

研究设计与样本

  • 系统综述与荟萃分析,系统检索PubMed、Web of Science、Cochrane Library和Embase数据库。
  • 最终纳入12项预测研究,共3474例患者进入荟萃分析。

主要发现

  • 良恶性前列腺癌鉴别:合并敏感度0.92(95%CI:0.83-0.97),合并特异度0.90(95%CI:0.68-0.97),合并AUC 0.96(95%CI:0.94-0.98)。
  • csPCa(Gleason评分≥7)诊断:合并敏感度0.83(95%CI:0.77-0.87),合并特异度0.73(95%CI:0.65-0.81),合并AUC 0.86(95%CI:0.83-0.89)。

关键数值与推荐

  • csPCa定义为Gleason评分≥7。
  • 采用敏感度、特异度和AUC评估基于机器学习MRI成像的诊断准确性。

临床怎么用

  • 基于机器学习的MRI成像可作为前列腺癌良恶性鉴别和csPCa检测的辅助诊断工具,但需结合临床综合判断。

局限性

  • 原文未报告纳入研究的具体异质性、偏倚风险及亚组分析结果。
  • 作者指出需要进一步深入研究验证 findings。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVE: This study aims to evaluate the diagnostic value of machine learning-based MRI imaging in differentiating benign and malignant prostate cancer and detecting clinically significant prostate cancer (csPCa, defined as Gleason score ≥7) using systematic review and meta-analysis methods. METHODS: Electronic databases (PubMed, Web of Science, Cochrane Library, and Embase) were systematically searched for predictive studies using machine learning-based MRI imaging for prostate cancer diagnosis. Sensitivity, specificity, and area under the curve (AUC) were used to assess the diagnostic accuracy of machine learning-based MRI imaging for both benign/malignant prostate cancer and csPCa. RESULTS: A total of 12 studies met the inclusion criteria, with 3474 patients included in the meta-analysis. Machine learning-based MRI imaging demonstrated good diagnostic value for both benign/malignant prostate cancer and csPCa. The pooled sensitivity and specificity for diagnosing benign/malignant prostate cancer were 0.92 (95% CI: 0.83-0.97) and 0.90 (95% CI: 0.68-0.97), respectively, with a combined AUC of 0.96 (95% CI: 0.94-0.98). For csPCa diagnosis, the pooled sensitivity and specificity were 0.83 (95% CI: 0.77-0.87) and 0.73 (95% CI: 0.65-0.81), respectively, with a combined AUC of 0.86 (95% CI: 0.83-0.89). CONCLUSION: Machine learning-based MRI imaging shows good diagnostic accuracy for both benign/malignant prostate cancer and csPCa. Further in-depth studies are needed to validate these findings.

来源:Prostate Cancer Prostatic Dis,查看原文。版权归原作者所有。

引用

  1. Zhao Y, Zhang L, Zhang S, et al. Machine learning-based MRI imaging for prostate cancer diagnosis: systematic review and meta-analysis. Prostate Cancer Prostatic Dis. 2025. PMID: 40721879. doi:10.1038/s41391-025-00997-2

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。