• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

使用生成式人工智能进行胶质瘤异柠檬酸脱氢酶突变预测的表型增强。

Phenotype augmentation using generative AI for isocitrate dehydrogenase mutation prediction in glioma.

作者信息

Jung Ha Kyung, Choi Changyong, Park Ji Eun, Park Seo Young, Lee Jae Ho, Kim Namkug, Kim Ho Sung

机构信息

Department of Radiology, Keimyung University Dongsan Hospital, Keimyung University School of Medicine, Daegu, Korea.

Department of Convergence Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea.

出版信息

Sci Rep. 2025 Aug 7;15(1):28913. doi: 10.1038/s41598-025-14477-z.

DOI:10.1038/s41598-025-14477-z
PMID:40775018
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC12331907/
Abstract

This study investigated the effects of feature augmentation, which uses generated images with specific imaging features, on the performance of isocitrate dehydrogenase (IDH) mutation prediction models in gliomas. A total of 598 patients were included from our institution (310 training, 152 internal test) and the Cancer Genome Atlas (136 external test). Score-based diffusion models were used to generate T2-weighted, FLAIR, and contrast-enhanced T1-weighted image triplets. Three neuroradiologists independently assessed visual Turing tests and various morphological features. Multivariable logistic regression models were developed using real images, random augmented data, and feature-augmented datasets. While random augmentation yielded models with AUCs comparable to real image-based models, it led to reduced specificity, particularly in the external test set (specificity: 83.2% vs. 73.0%, P = .013). In contrast, feature-augmented models maintained stable diagnostic performance; however, when more than 70% of training images included synthetic T2-FLAIR mismatch signs, AUC decreased in the external test set (AUC: 0.905-0.906 for ≤ 70%; 0.902-0.876 for ≥ 80%). These findings highlight the value of phenotype-specific augmentation for IDH prediction, while emphasizing the need to optimize augmentation proportion to avoid performance degradation.

摘要

本研究调查了特征增强(使用具有特定成像特征的生成图像)对神经胶质瘤中异柠檬酸脱氢酶(IDH)突变预测模型性能的影响。我们机构共纳入598例患者(310例用于训练,152例用于内部测试)以及癌症基因组图谱(136例用于外部测试)。基于分数的扩散模型用于生成T2加权、液体衰减反转恢复序列(FLAIR)和对比增强T1加权图像三联体。三名神经放射科医生独立评估视觉图灵测试和各种形态学特征。使用真实图像、随机增强数据和特征增强数据集开发多变量逻辑回归模型。虽然随机增强产生的模型的曲线下面积(AUC)与基于真实图像的模型相当,但它导致特异性降低,尤其是在外部测试集中(特异性:83.2%对73.0%,P = 0.013)。相比之下,特征增强模型保持了稳定的诊断性能;然而,当超过70%的训练图像包含合成的T2-FLAIR不匹配征象时,外部测试集中的AUC下降(AUC:≤70%时为0.905至0.906;≥80%时为0.902至0.876)。这些发现突出了表型特异性增强对IDH预测的价值,同时强调需要优化增强比例以避免性能下降。

https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/8b02e3daaf73/41598_2025_14477_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/8bfa6da85173/41598_2025_14477_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/4b7c47beb015/41598_2025_14477_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/8b02e3daaf73/41598_2025_14477_Fig3_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/8bfa6da85173/41598_2025_14477_Fig1_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/4b7c47beb015/41598_2025_14477_Fig2_HTML.jpg
https://cdn.ncbi.nlm.nih.gov/pmc/blobs/469e/12331907/8b02e3daaf73/41598_2025_14477_Fig3_HTML.jpg

相似文献

1
Phenotype augmentation using generative AI for isocitrate dehydrogenase mutation prediction in glioma.使用生成式人工智能进行胶质瘤异柠檬酸脱氢酶突变预测的表型增强。
Sci Rep. 2025 Aug 7;15(1):28913. doi: 10.1038/s41598-025-14477-z.
2
Nonenhancing Margin and Pial Invasion in Magnetic Resonance Imaging can Predict Isocitrate Dehydrogenase Status in Glioma Patients.磁共振成像中的无强化边缘和软脑膜侵犯可预测胶质瘤患者的异柠檬酸脱氢酶状态。
World Neurosurg. 2025 Mar;195:123624. doi: 10.1016/j.wneu.2024.123624. Epub 2025 Jan 26.
3
Using partially shared radiomics features to simultaneously identify isocitrate dehydrogenase mutation status and epilepsy in glioma patients from MRI images.利用部分共享的影像组学特征从磁共振成像(MRI)图像中同时识别胶质瘤患者的异柠檬酸脱氢酶突变状态和癫痫情况。
Sci Rep. 2025 Jan 28;15(1):3591. doi: 10.1038/s41598-025-87778-y.
4
MRI-based habitat radiomics for preoperatively predicting IDH status in gliomas.基于磁共振成像的肿瘤栖息地放射组学用于术前预测胶质瘤中的异柠檬酸脱氢酶状态
Neurosurg Focus. 2025 Aug 1;59(2):E4. doi: 10.3171/2025.5.FOCUS25135.
5
Radiogenomic association between the T2-FLAIR mismatch sign and IDH mutation status in adult patients with lower-grade gliomas: an updated systematic review and meta-analysis.成人低级别胶质瘤患者 T2-FLAIR 不匹配征象与 IDH 突变状态的放射基因组关联:一项更新的系统评价和荟萃分析。
Eur Radiol. 2022 Aug;32(8):5339-5352. doi: 10.1007/s00330-022-08607-8. Epub 2022 Feb 15.
6
Predicting p53 Status in IDH-Mutant Gliomas Using MRI-Based Radiomic Model.使用基于MRI的放射组学模型预测IDH突变型胶质瘤中的p53状态
Cancer Med. 2025 Aug;14(15):e71063. doi: 10.1002/cam4.71063.
7
Diffusion MRI is superior to quantitative T2-FLAIR mismatch in predicting molecular subtypes of human non-enhancing gliomas.在预测人类非强化型胶质瘤的分子亚型方面,扩散加权磁共振成像优于定量T2液体衰减反转恢复序列不匹配。
Neuroradiology. 2024 Dec;66(12):2153-2162. doi: 10.1007/s00234-024-03475-z. Epub 2024 Oct 8.
8
Presence of Fragmented Intratumoral Thrombosed Microvasculature in the Necrotic and Peri-Necrotic Regions on SWI Differentiates IDH Wild-Type Glioblastoma From IDH Mutant Grade 4 Astrocytoma.磁敏感加权成像(SWI)上坏死及坏死周边区域存在瘤内破碎血栓形成的微血管,可将异柠檬酸脱氢酶(IDH)野生型胶质母细胞瘤与IDH突变型4级星形细胞瘤区分开来。
J Magn Reson Imaging. 2025 Jul;62(1):258-270. doi: 10.1002/jmri.29695. Epub 2025 Jan 9.
9
Clinicopathological and radiological characteristics of false-positive and false-negative results in T2-FLAIR mismatch sign of IDH-mutated gliomas.IDH 突变型胶质瘤 T2-FLAIR 错配征象中假阳性和假阴性结果的临床病理及影像学特征。
Clin Neurol Neurosurg. 2024 Nov;246:108579. doi: 10.1016/j.clineuro.2024.108579. Epub 2024 Oct 1.
10
Predicting the molecular subtypes of 2021 WHO grade 4 glioma by a multiparametric MRI-based machine learning model.基于多参数磁共振成像的机器学习模型预测2021年世界卫生组织4级胶质瘤的分子亚型
BMC Cancer. 2025 Jul 14;25(1):1171. doi: 10.1186/s12885-025-14529-7.

本文引用的文献

1
Maximum Resection of Noncontrast-enhanced Tumor at MRI Is a Favorable Prognostic Factor in IDH Wild-Type Glioblastoma.磁共振成像(MRI)上非强化肿瘤的最大程度切除是异柠檬酸脱氢酶(IDH)野生型胶质母细胞瘤的一个有利预后因素。
Radiology. 2025 May;315(2):e241393. doi: 10.1148/radiol.241393.
2
Image-Based Generative Artificial Intelligence in Radiology: Comprehensive Updates.基于图像的放射学生成式人工智能:全面更新。
Korean J Radiol. 2024 Nov;25(11):959-981. doi: 10.3348/kjr.2024.0392.
3
Association between dichotomized VASARI feature and overall survival in glioblastoma patients: a single-institution propensity score matching analysis.
二分类 VASARI 特征与胶质母细胞瘤患者总生存期的相关性:单中心倾向评分匹配分析。
Cancer Imaging. 2024 Aug 18;24(1):109. doi: 10.1186/s40644-024-00754-z.
4
Updated Primer on Generative Artificial Intelligence and Large Language Models in Medical Imaging for Medical Professionals.医学专业人员医学影像生成式人工智能和大型语言模型更新基础篇。
Korean J Radiol. 2024 Mar;25(3):224-242. doi: 10.3348/kjr.2023.0818.
5
Generative AI in glioma: Ensuring diversity in training image phenotypes to improve diagnostic performance for IDH mutation prediction.生成式人工智能在脑胶质瘤中的应用:通过确保训练图像表型的多样性,提高 IDH 突变预测的诊断性能。
Neuro Oncol. 2024 Jun 3;26(6):1124-1135. doi: 10.1093/neuonc/noae012.
6
RANO 2.0: Update to the Response Assessment in Neuro-Oncology Criteria for High- and Low-Grade Gliomas in Adults. RANO 2.0:成人高级别和低级别胶质瘤反应评估标准更新。
J Clin Oncol. 2023 Nov 20;41(33):5187-5199. doi: 10.1200/JCO.23.01059. Epub 2023 Sep 29.
7
The 2021 WHO Classification for Gliomas and Implications on Imaging Diagnosis: Part 1-Key Points of the Fifth Edition and Summary of Imaging Findings on Adult-Type Diffuse Gliomas.2021 年世界卫生组织(WHO)脑肿瘤分类及其对影像诊断的影响:第五版要点——成人弥漫性胶质瘤的影像学表现概述。
J Magn Reson Imaging. 2023 Sep;58(3):677-689. doi: 10.1002/jmri.28743. Epub 2023 Apr 17.
8
Preoperative and Noninvasive Prediction of Gliomas Histopathological Grades and IDH Molecular Types Using Multiple MRI Characteristics.利用多种MRI特征对胶质瘤组织病理学分级和IDH分子类型进行术前及无创预测
Front Oncol. 2022 May 27;12:873839. doi: 10.3389/fonc.2022.873839. eCollection 2022.
9
Conventional MRI features can predict the molecular subtype of adult grade 2-3 intracranial diffuse gliomas.常规 MRI 特征可预测成人 2-3 级颅内弥漫性胶质瘤的分子亚型。
Neuroradiology. 2022 Dec;64(12):2295-2305. doi: 10.1007/s00234-022-02975-0. Epub 2022 May 24.
10
Utilizing Synthetic Nodules for Improving Nodule Detection in Chest Radiographs.利用合成结节来提高胸部 X 光片中的结节检测。
J Digit Imaging. 2022 Aug;35(4):1061-1068. doi: 10.1007/s10278-022-00608-9. Epub 2022 Mar 18.