文献速递神经技术与安全CTMR

人工智能检测颅内出血与慢性脑微出血的系统综述与汇总分析

Accuracy of artificial intelligence for the detection of intracranial hemorrhage and chronic cerebral microbleeds: a systematic review and pooled analysis.
PubMed 文献 · Radiol Med · 2022-08-13

该研究为PROSPERO注册、PRISMA compliant的系统综述,汇总了MEDLINE和EMBASE中开发或测试AI算法用于非增强CT/MRI检测颅内出血(ICH)及MRI检测慢性脑微出血(CMBs)的研究。共纳入40项ICH检测研究和19项CMBs检测研究,汇总敏感度、特异度和准确率分别为ICH 92.06%、93.54%、93.46%,CMBs 91.6%、93.9%、92.7%。结论认为AI工具平均性能高、达到专家级准确率,可改善工作流程并作为检测此类病变的安全网。

要点

  • AI在非增强CT/MRI上检测ICH的汇总敏感度92.06%、特异度93.54%、准确率93.46%。
  • AI在MRI上检测CMBs的汇总敏感度91.6%、特异度93.9%、准确率92.7%。
  • AI工具平均性能达专家级,可辅助检出ICH和CMBs并改善流程。

核心结论

  • 过去十年已开发众多AI驱动软件工具,平均对ICH和CMBs诊断具有高性能和专家级准确率。
  • 将这些工具用于临床实践可能改善工作流程,并作为检测此类病变的安全网。

研究设计与样本

  • PROSPERO注册(CRD42021246848)、PRISMA compliant的系统综述与汇总分析。
  • 检索MEDLINE和EMBASE,纳入所有开发或测试AI算法用于非增强CT/MRI检测ICH及MRI检测CMBs的研究。
  • 共纳入40项ICH检测研究(非增强CT/MRI)和19项CMBs检测研究(MRI)。

诊断标准与关键数值

  • ICH检测(非增强CT/MRI):汇总敏感度92.06%、特异度93.54%、准确率93.46%。
  • CMBs检测(MRI):汇总敏感度91.6%、特异度93.9%、准确率92.7%。

临床应用与阅片要点

  • AI工具可作为ICH和CMBs检测的辅助手段,用于改善工作流程并作为安全网。
  • 算法开发中需应对标注数据集构建工作量大、影像为容积性检查、算法微调及减少假阳性等挑战。

局限性

  • 原文指出算法开发面临创建大型、标注且平衡数据集的繁重工作、影像检查的容积性质、算法微调及减少假阳性等挑战。

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

BACKGROUND: Artificial intelligence (AI)-driven software has been developed and become commercially available within the past few years for the detection of intracranial hemorrhage (ICH) and chronic cerebral microbleeds (CMBs). However, there is currently no systematic review that summarizes all of these tools or provides pooled estimates of their performance.

METHODS: In this PROSPERO-registered, PRISMA compliant systematic review, we sought to compile and review all MEDLINE and EMBASE published studies that have developed and/or tested AI algorithms for ICH detection on non-contrast CT scans (NCCTs) or MRI scans and CMBs detection on MRI scans.

RESULTS: In total, 40 studies described AI algorithms for ICH detection in NCCTs/MRIs and 19 for CMBs detection in MRIs. The overall sensitivity, specificity, and accuracy were 92.06%, 93.54%, and 93.46%, respectively, for ICH detection and 91.6%, 93.9%, and 92.7% for CMBs detection. Some of the challenges encountered in the development of these algorithms include the laborious work of creating large, labeled and balanced datasets, the volumetric nature of the imaging examinations, the fine tuning of the algorithms, and the reduction in false positives.

CONCLUSIONS: Numerous AI-driven software tools have been developed over the last decade. On average, they are characterized by high performance and expert-level accuracy for the diagnosis of ICH and CMBs. As a result, implementing these tools in clinical practice may improve workflow and act as a failsafe for the detection of such lesions. REGISTRATION-URL: https://www.crd.york.ac.uk/prospero/ Unique Identifier: CRD42021246848.

来源:Radiol Med,查看原文。版权归原作者所有。

引用

  1. Matsoukas, Scaggiante, Schuldt, et al. Accuracy of artificial intelligence for the detection of intracranial hemorrhage and chronic cerebral microbleeds: a systematic review and pooled analysis. Radiol Med. 2022. PMID: 35962888. doi:10.1007/s11547-022-01530-4

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。