北京天坛医院神经外科教授宫剑团队近期取得重要工作进展，他们研究利用人工智能和磁共振成像特征推进髓母细胞瘤术前非侵入性分子亚群预测。相关研究成果2024年6月27日在线发表于《癌细胞》杂志上。

据介绍，髓母细胞瘤的全球研究由于普遍无法进行分子亚群检测和缺乏数据而受到阻碍。

为了弥补这一差距，研究人员建立了一个国际分子特征数据库，包括来自中国和美国13个中心的934名髓母细胞瘤患者。研究人员展示了基于图像的机器学习策略在髓母细胞瘤的临床管理中为无创、术前和低成本的分子亚群预测中创造一种替代途径。这一稳健的验证策略（包括交叉验证、外部验证和连续验证）证明了该模型作为可推广的分子诊断分类器的有效性。详细的MRI特征分析通过精密的放射学镜头补充了对髓母细胞瘤的认识。

此外，东亚和北美亚群之间的比较突出了重要的管理意义。研究人员公开了包括MRI特征、临床病理特征、治疗变量和生存数据这一全面的数据集，以推进全球髓母细胞瘤研究进展。

附：英文原文

Title: Advancing presurgical non-invasive molecular subgroup prediction in medulloblastoma using artificial intelligence and MRI signatures

Author: Yan-Ran (Joyce) Wang, Pengcheng Wang, Zihan Yan, Quan Zhou, Fatma Gunturkun, Peng Li, Yanshen Hu, Wei Emma Wu, Kankan Zhao, Michael Zhang, Haoyi Lv, Lehao Fu, Jiajie Jin, Qing Du, Haoyu Wang, Kun Chen, Liangqiong Qu, Keldon Lin, Michael Iv, Hao Wang, Xiaoyan Sun, Hannes Vogel, Summer Han, Lu Tian, Feng Wu, Jian Gong

Issue&Volume: 2024-06-27

Abstract: Global investigation of medulloblastoma has been hindered by the widespread inaccessibilityof molecular subgroup testing and paucity of data. To bridge this gap, we establishedan international molecularly characterized database encompassing 934 medulloblastomapatients from thirteen centers across China and the United States. We demonstratehow image-based machine learning strategies have the potential to create an alternativepathway for non-invasive, presurgical, and low-cost molecular subgroup predictionin the clinical management of medulloblastoma. Our robust validation strategies—includingcross-validation, external validation, and consecutive validation—demonstrate themodel’s efficacy as a generalizable molecular diagnosis classifier. The detailed analysisof MRI characteristics replenishes the understanding of medulloblastoma through anuanced radiographic lens. Additionally, comparisons between East Asia and North Americasubsets highlight critical management implications. We made this comprehensive dataset,which includes MRI signatures, clinicopathological features, treatment variables,and survival data, publicly available to advance global medulloblastoma research.

DOI: 10.1016/j.ccell.2024.06.002

Source: https://www.cell.com/cancer-cell/abstract/S1535-6108(24)00227-7