Unsupervised machine learning using K-means identifies radiomic subgroups of pediatric low-grade gliomas that correlate with key molecular markers.
Unsupervised machine learning using K-means identifies radiomic subgroups of pediatric low-grade gliomas that correlate with key molecular markers.
复制标题
DOI：
10.1016/j.neo.2022.100869
复制
发表时间：
2023-02
期刊：
影响因子：
--
通讯作者：
Nabavizadeh A
中科院分区：
文献类型：
--
作者：
Haldar D;Kazerooni AF;Arif S;Familiar A;Madhogarhia R;Khalili N;Bagheri S;Anderson H;Shaikh IS;Mahtabfar A;Kim MC;Tu W;Ware J;Vossough A;Davatzikos C;Storm PB;Resnick A;Nabavizadeh A
尽管在儿童低级别胶质瘤（pLGG）的分子和组织病理学特征方面取得了进展，但在具有相似分类的肿瘤之间仍然存在显著的表型异质性。我们假设基于放射组学特征的无监督机器学习方法可以揭示不同的pLGG成像亚型。收集了157例pLGGs患者的多参数MR图像（T1增强前和增强后，T2和T2 FLAIR），并从肿瘤区域提取了881个定量放射学特征。在应用主成分分析（PCA）进行特征降维后，使用K均值进行聚类。从PedCBioportal获得分子和人口统计学数据，并在成像亚型之间进行比较。K-means确定了三种不同的基于成像的亚型。不同亚型的BRAF基因突变频率不同（p < 0.05），BRAF基因表达也不同（p<0.05）。还发现年龄（p < 0.05）、肿瘤位置（p < 0.01）和肿瘤组织学（p < 0.0001）在成像亚型之间存在显著差异。在这项探索性工作中，发现基于放射组学特征的pLGG聚类识别了与重要分子标志物和人口统计学细节相关的不同的基于成像的亚型。这一发现支持了放射组学数据的结合可以增强我们更好地表征pLGG的能力的观点。
Despite advancements in molecular and histopathologic characterization of pediatric low-grade gliomas (pLGGs), there remains significant phenotypic heterogeneity among tumors with similar categorizations. We hypothesized that an unsupervised machine learning approach based on radiomic features may reveal distinct pLGG imaging subtypes. Multi-parametric MR images (T1 pre- and post-contrast, T2, and T2 FLAIR) from 157 patients with pLGGs were collected and 881 quantitative radiomic features were extracted from tumorous region. Clustering was performed using K-means after applying principal component analysis (PCA) for feature dimensionality reduction. Molecular and demographic data was obtained from the PedCBioportal and compared between imaging subtypes. K-means identified three distinct imaging-based subtypes. Subtypes differed in mutational frequencies of BRAF (p < 0.05) as well as the gene expression of BRAF (p<0.05). It was also found that age (p < 0.05), tumor location (p < 0.01), and tumor histology (p < 0.0001) differed significantly between the imaging subtypes. In this exploratory work, it was found that clustering of pLGGs based on radiomic features identifies distinct, imaging-based subtypes that correlate with important molecular markers and demographic details. This finding supports the notion that incorporation of radiomic data could augment our ability to better characterize pLGGs.
登录
查看更多内容
影响因子：
7.3
作者：
Gao J;Aksoy BA;Dogrusoz U;Dresdner G;Gross B;Sumer SO;Sun Y;Jacobsen A;Sinha R;Larsson E;Cerami E;Sander C;Schultz N
通讯作者：
Schultz N
影响因子：
4.8
作者：
Rohlfing, Torsten;Zahr, Natalie M.;Sullivan, Edith V.;Pfefferbaum, Adolf
通讯作者：
Pfefferbaum, Adolf
影响因子：
5.2
作者：
Fathi Kazerooni A;Bagley SJ;Akbari H;Saxena S;Bagheri S;Guo J;Chawla S;Nabavizadeh A;Mohan S;Bakas S;Davatzikos C;Nasrallah MP
通讯作者：
Nasrallah MP
影响因子：
45.3
作者：
Lassaletta, Alvaro;Zapotocky, Michal;Tabori, Uri
通讯作者：
Tabori, Uri
对进行性小儿低度神经胶质瘤患者的曲敏尼治疗的反应。
影响因子：
3.9
作者：
Selt F;van Tilburg CM;Bison B;Sievers P;Harting I;Ecker J;Pajtler KW;Sahm F;Bahr A;Simon M;Jones DTW;Well L;Mautner VF;Capper D;Hernáiz Driever P;Gnekow A;Pfister SM;Witt O;Milde T
通讯作者：
Milde T