文献速递腹部CTMR

机器学习算法鉴别肝脏黏液性囊性肿瘤与良性肝囊肿的外部验证

External validation of a machine learning based algorithm to differentiate hepatic mucinous cystic neoplasms from benign hepatic cysts.
PubMed 文献 · Abdom Radiol (NY) · 2023-04-13

本研究在多机构回顾性队列中外部验证Hardie等提出的3特征分类算法,用于无创鉴别肝脏黏液性囊性肿瘤(MCN)与良性肝囊肿(BHC)。共纳入159例病理确诊患者,算法准确度达98.1%(95%CI 94.6%-99.6%),AUC为0.911,放射科医师间分类一致性几乎完美(Fleiss Kappa 0.840),提示该算法可作为临床决策支持工具。

要点

  • 3特征算法在外部验证中准确度98.1%,AUC 0.911,可辅助鉴别MCN与BHC。
  • 放射科医师间分类一致性几乎完美(Fleiss Kappa 0.840),特征可重复。
  • 算法阳性预测值100.0%,阴性预测值97.9%,但MCN样本量较小(17例)。

核心结论

  • 该3特征分类算法在外部多机构验证队列中表现出与原始报道相似的高诊断准确度,可作为临床决策支持工具。

研究设计与样本

  • 多机构回顾性研究,纳入2005年1月至2022年3月病理确诊为MCN或BHC的囊性肝病变患者159例。
  • 5名读者(2名放射科医师、3名非放射科医师)独立回顾组织取样前的增强CT或MRI,应用Hardie等描述的3特征分类算法。
  • 以病理结果为参考标准,计算算法准确度、阳性预测值、阴性预测值和AUC,并用Fleiss Kappa评估读者间一致性。

诊断标准与关键数值

  • 算法鉴别MCN与BHC的准确度为98.1%(95%CI 94.6%-99.6%)。
  • 阳性预测值为100.0%(95%CI 76.8%-100.0%),阴性预测值为97.9%(95%CI 94.1%-99.6%)。
  • AUC为0.911(95%CI 0.818-1.000)。
  • 放射科医师间分类一致性Fleiss Kappa为0.840(p<0.001)。

临床应用与阅片要点

  • 该3特征算法易于快速应用,且特征在放射科医师间可重复,适合作为日常阅片辅助工具。
  • 应用时需注意MCN患病率较低(10.7%),阳性预测值置信区间较宽。

局限性

  • 回顾性研究设计。
  • MCN样本量较小(17例),阳性预测值置信区间宽。

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 externally validate an algorithm for non-invasive differentiation of hepatic mucinous cystic neoplasms (MCN) from benign hepatic cysts (BHC), which differ in management.

METHODS: Patients with cystic liver lesions pathologically confirmed as MCN or BHC between January 2005 and March 2022 from multiple institutions were retrospectively included. Five readers (2 radiologists, 3 non-radiologist physicians) independently reviewed contrast-enhanced CT or MRI examinations before tissue sampling and applied the 3-feature classification algorithm described by Hardie et al. to differentiate between MCN and BHC, which had a reported accuracy of 93.5%. The classification was then compared to the pathology results. Interreader agreement between readers across different levels of experience was evaluated with Fleiss' Kappa.

RESULTS: The final cohort included 159 patients, median age of 62 years (IQR [52.0, 70.0]), 66.7% female (106). Of all patients, 89.3% (142) had BHC, and the remaining 10.7% (17) had MCN on pathology. Agreement for class designation between the radiologists was almost perfect (Fleiss' Kappa 0.840, p < 0.001). The algorithm had an accuracy of 98.1% (95% CI [94.6%, 99.6%]), a positive predictive value of 100.0% (95% CI [76.8%, 100.0%]), a negative predictive value of 97.9% (95% CI [94.1%, 99.6%]), and an area under the receiver operator characteristic curve (AUC) of 0.911 (95% CI [0.818, 1.000]).

CONCLUSION: The evaluated algorithm showed similarly high diagnostic accuracy in our external, multi-institutional validation cohort. This 3-feature algorithm is easily and rapidly applied and its features are reproducible among radiologists, showing promise as a clinical decision support tool.

来源:Abdom Radiol (NY),查看原文。版权归原作者所有。

引用

  1. Furtado, Badenes-Romero, Hesami, et al. External validation of a machine learning based algorithm to differentiate hepatic mucinous cystic neoplasms from benign hepatic cysts. Abdom Radiol (NY). 2023. PMID: 37055585. doi:10.1007/s00261-023-03907-z

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。