机器学习模型在对比增强CT图像中用于胰腺导管腺癌预诊断与诊断的准确性:系统综述与荟萃分析
本系统综述与荟萃分析纳入9项研究共12788例受试者(其中3997例进入荟萃分析),评估机器学习模型在对比增强CT图像中检测胰腺导管腺癌的诊断能力。基于CT的AI模型准确度为88.7%(95%CI 87.7%-89.7%),敏感度87.9%(95%CI 82.9%-91.6%),特异度92.2%(95%CI 86.8%-95.5%)。作者认为尽管敏感度和特异度结果良好,但纳入研究质量较低,限制了确定性结论,需进一步研究验证后方可广泛采用。
要点
- 基于对比增强CT的机器学习模型检测胰腺导管腺癌:敏感度87.9%、特异度92.2%、准确度88.7%。
- 六项影像组学研究平均RQS评分为17.83分,九项机器学习方法平均CLAIM评分为30.55分。
- 作者指出纳入研究质量较低,限制确定性结论,需进一步研究验证。
核心结论
- 本研究表明基于对比增强CT的机器学习模型检测胰腺导管腺癌具有较好的敏感度(87.9%)和特异度(92.2%),但纳入研究质量较低,限制了确定性结论,需进一步研究验证这些模型后方可广泛采用。
研究设计与样本
- 系统综述与荟萃分析,检索PubMed、Cochrane对照试验中心注册库和Embase(无限制),纳入经组织病理学确诊的成人胰腺导管腺癌研究,最终纳入9项研究、评估12788例受试者,其中3997例进入荟萃分析。
- 采用QUADAS-2评估偏倚风险,采用影像组学质量评分(RQS)和医学影像人工智能清单(CLAIM)评估方法学质量,采用双变量随机效应模型对敏感度和特异度进行荟萃分析,采用I²值和亚组分析评估异质性。
诊断标准与关键数值
- 基于CT扫描的AI模型检测胰腺导管腺癌:准确度88.7%(95%CI 87.7%-89.7%)、敏感度87.9%(95%CI 82.9%-91.6%)、特异度92.2%(95%CI 86.8%-95.5%)。
- 六项影像组学研究的平均RQS评分为17.83分。
- 九项机器学习方法的平均CLAIM评分为30.55分。
临床应用与阅片要点
- 原文结论指出,尽管基于对比增强CT的机器学习模型检测胰腺导管腺癌的敏感度和特异度结果良好,但纳入研究质量较低,限制了确定性结论,需进一步研究验证这些模型后方可广泛采用。
局限性
- 作者指出纳入研究质量较低,限制了确定性结论。
- 作者指出需进一步研究验证这些模型后方可广泛采用。
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)。
英文原文(摘要)· 点击展开
PURPOSE: To evaluate the diagnostic ability and methodological quality of ML models in detecting Pancreatic Ductal Adenocarcinoma (PDAC) in Contrast CT images.
METHOD: Included studies assessed adults diagnosed with PDAC, confirmed by histopathology. Metrics of tests were interpreted by ML algorithms. Studies provided data on sensitivity and specificity. Studies that did not meet the inclusion criteria, segmentation-focused studies, multiple classifiers or non-diagnostic studies were excluded. PubMed, Cochrane Central Register of Controlled Trials, and Embase were searched without restrictions. Risk of bias was assessed using QUADAS-2, methodological quality was evaluated using Radiomics Quality Score (RQS) and a Checklist for AI in Medical Imaging (CLAIM). Bivariate random-effects models were used for meta-analysis of sensitivity and specificity, I2 values and subgroup analysis used to assess heterogeneity.
RESULTS: Nine studies were included and 12,788 participants were evaluated, of which 3,997 were included in the meta-analysis. AI models based on CT scans showed an accuracy of 88.7% (IC 95%, 87.7%-89.7%), sensitivity of 87.9% (95% CI, 82.9%-91.6%), and specificity of 92.2% (95% CI, 86.8%-95.5%). The average score of six radiomics studies was 17.83 RQS points. Nine ML methods had an average CLAIM score of 30.55 points.
CONCLUSIONS: Our study is the first to quantitatively interpret various independent research, offering insights for clinical application. Despite favorable sensitivity and specificity results, the studies were of low quality, limiting definitive conclusions. Further research is necessary to validate these models before widespread adoption.
引用
- Lopes Costa GL, Tasca Petroski G, Machado LG, et al. Accuracy of machine learning models for pre-diagnosis and diagnosis of pancreatic ductal adenocarcinoma in contrast-CT images: a systematic review and meta-analysis. Abdom Radiol (NY). 2024. PMID: 39720966. doi:10.1007/s00261-024-04771-1
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。