人工智能与影像组学用于儿童脑肿瘤分类及分子特征描述:系统综述
该系统综述纳入24项研究,评估人工智能(含影像组学/机器学习与深度学习)在儿童脑肿瘤影像诊断分类、分子分型中的性能与方法学质量。多数研究基于MRI(96%)和影像组学机器学习模型(88%),诊断任务AUC范围为0.73至0.98(中位数0.91),但仅3项研究(12.5%)在独立队列进行外部验证。作者结论认为人工智能与影像组学诊断准确性高,但缺乏前瞻性设计及外部验证率极低限制了泛化性,需要标准化多中心研究支持更广泛临床采用。
要点
- 儿童脑肿瘤AI诊断任务报告AUC范围为0.73至0.98,中位数0.91。
- 仅3项研究(12.5%)在独立队列进行外部验证。
- 19项研究(79%)经NOS评为低偏倚风险(评分8-9分)。
核心结论
- 人工智能与影像组学对儿童脑肿瘤特征描述表现出高诊断准确性,但缺乏前瞻性设计且外部验证率极低限制了泛化性,需要标准化多中心研究支持更广泛临床采用。
研究设计与样本
- 系统综述按照PRISMA 2020指南进行,检索PubMed、Scopus和Web of Science(2010年至2025年)。
- 纳入标准为年龄0至18岁脑肿瘤患者、对神经影像应用人工智能进行诊断分类、分子特征描述或整体图像处理任务并报告性能指标的研究。
- 从638条记录中纳入24项研究,方法学质量采用纽卡斯尔-渥太华量表(NOS)和医学影像人工智能清单(CLAIM)评估。
诊断标准与关键数值
- 诊断任务报告的AUC范围为0.73至0.98,中位数0.91。
- 基于NOS,19项研究(79%)评为低偏倚风险(评分8-9分)。
- 仅3项研究(12.5%)在独立队列进行外部验证。
- 多数研究使用MRI(96%)和基于影像组学的机器学习模型(88%),较小比例采用深度学习(29%)。
- 主要诊断任务为肿瘤分类(50%)和分子亚型预测(33%)。
临床应用与阅片要点
- 原文未据此给出推荐。
局限性
- 缺乏前瞻性设计。
- 外部验证率极低(仅3项研究,12.5%在独立队列进行外部验证),限制了泛化性。
- 方法学存在变异性且缺乏标准化。
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) and radiomics are increasingly applied in pediatric neuroradiology to enhance diagnostic precision. However, their clinical implementation remains limited due to methodological variability and lack of standardization.
OBJECTIVE: To systematically evaluate the diagnostic applications, performance, and methodological quality of artificial intelligence models, including both radiomics/machine learning and deep learning approaches, in pediatric brain tumor imaging.
METHODS: A systematic review was conducted in accordance with PRISMA 2020 guidelines, searching PubMed, Scopus, and Web of Science (2010-2025). Eligible studies included patients aged 0-18 years with brain tumors, applied AI to neuroimaging for diagnostic classification, molecular characterization, or integral image processing tasks, and reported performance metrics. Methodological quality was assessed using the Newcastle-Ottawa Scale (NOS) and the Checklist for Artificial Intelligence in Medical Imaging (CLAIM).
RESULTS: From 638 records, 24 studies were included. Most studies used MRI (96%) and machine learning models based on radiomics (88%), with a smaller proportion employing deep learning (29%). The primary diagnostic tasks were tumor classification (50%) and molecular subtype prediction (33%). Reported AUCs for diagnostic tasks ranged from 0.73 to 0.98 (median: 0.91). Based on the NOS, 19 studies (79%) were rated as low risk of bias (scores 8-9), though only 3 studies (12.5%) performed external validation on independent cohorts.
CONCLUSION: AI and radiomics demonstrate high diagnostic accuracy for pediatric brain tumor characterization. Nonetheless, the lack of prospective design and critically low rate of external validation limits generalizability. Standardized, multicenter studies are needed to support broader clinical adoption.
引用
- Reyes JS, Estupiñan-Pepinosa DF, Leal-Giraldo SI, et al. Artificial intelligence and radiomics for pediatric brain tumor classification and molecular characterization: a systematic review. Neuroradiology. 2026. PMID: 41894040. doi:10.1007/s00234-026-03976-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。