Information theory approaches to improve glioma diagnostic workflows in surgical neuropathology.
Information theory approaches to improve glioma diagnostic workflows in surgical neuropathology.
复制标题
DOI：
10.1111/bpa.13050
复制
发表时间：
2022-09
期刊：
影响因子：
--
通讯作者：
中科院分区：
文献类型：
--
作者：
资源紧张的医疗保健生态系统通常难以采用世界卫生组织（WHO）关于中枢神经系统（CNS）肿瘤分类的建议。强大的临床诊断辅助工具的产生和简单的解决方案的进步，以告知外科神经病理学的投资策略，将改善这些环境中的患者护理。我们在脑癌模拟模型和真实的世界数据集上使用简单的信息论计算来比较临床、组织学、免疫组织化学和分子信息的贡献。生成图像噪声测定以比较不同图像分割方法在从数字载玻片获得的H&E和Olig 2染色图像中的效率。生成自动可调图像分析工作流程，并与神经病理学家进行p53阳性定量比较。最后，提取的细胞核特征密度、p53阳性定量和组合的ATRX/年龄特征用于生成IDH突变型肿瘤中1 p/19 q共缺失的预测模型。信息理论计算可以在开放获取平台上进行，并提供对诊断生物标志物之间的线性和非线性关联的重要见解。年龄、p53和ATRX状态对IDH突变型肿瘤的诊断有重要意义。预测模型可能有助于减少荧光原位杂交（FISH）检测的假阳性1 p/19 q共缺失。我们认为，这种方法为IDH突变型肿瘤的cIMPACT‐NOW工作流程建议提供了改进，并为未来的资源和测试分配提供了框架。不同的聚类模式与临床，组织学，免疫组化和分子信息对脑癌人群的模拟和信息增益在神经胶质瘤模拟模型的临床决策。A1-A4中显示了通过主成分分析进行的近似性降低，B1-B4中显示了通过t-随机邻居嵌入进行的近似性降低，每个图的顶部都描绘了特征。每种颜色代表WHO分类方案中的唯一诊断。(A1和B1）仅具有临床特征的模糊性降低仅证明了少数可见的聚类。(A2和B2）用组织学来描述临床病史产生了PCA中的清晰分层和t-SNE的离散聚类的开始。(A3和B3）包括免疫组织化学数据提高了辨别簇的能力。(A4和B4）附加的分子特征不显著改善聚类。(C)在胶质瘤模拟模型中获得了所有临床、组织学、免疫组化和分子特征的个体信息。年龄、部位和Ki 67是诊断所需信息最多的特征。(D)临床、组织学、免疫组化和分子信息的信息增益。组织学为诊断提供了大部分必要的信息。将临床数据与组织学和免疫组织化学相结合提供了超过95%的必要信息，而添加分子数据提供了最小的信息增益。(E)对IDH 1/2突变的肿瘤进行子集化，并将X轴上特征的互信息定量为Y轴上的%信息。对1 p19 q状态的数据调用条件信息函数。
Resource‐strained healthcare ecosystems often struggle with the adoption of the World Health Organization (WHO) recommendations for the classification of central nervous system (CNS) tumors. The generation of robust clinical diagnostic aids and the advancement of simple solutions to inform investment strategies in surgical neuropathology would improve patient care in these settings. We used simple information theory calculations on a brain cancer simulation model and real‐world data sets to compare contributions of clinical, histologic, immunohistochemical, and molecular information. An image noise assay was generated to compare the efficiencies of different image segmentation methods in H&E and Olig2 stained images obtained from digital slides. An auto‐adjustable image analysis workflow was generated and compared with neuropathologists for p53 positivity quantification. Finally, the density of extracted features of the nuclei, p53 positivity quantification, and combined ATRX/age feature was used to generate a predictive model for 1p/19q codeletion in IDH‐mutant tumors. Information theory calculations can be performed on open access platforms and provide significant insight into linear and nonlinear associations between diagnostic biomarkers. Age, p53, and ATRX status have significant information for the diagnosis of IDH‐mutant tumors. The predictive models may facilitate the reduction of false‐positive 1p/19q codeletion by fluorescence in situ hybridization (FISH) testing. We posit that this approach provides an improvement on the cIMPACT‐NOW workflow recommendations for IDH‐mutant tumors and a framework for future resource and testing allocation. Different clustering patterns with clinical, histologic, immunohistochemical, and molecular information on the brain cancer population simulation and information gain in the glioma simulation model for clinical decision‐making. Dimensionality reduction by principal component analysis is shown in A1–A4, and by t‐stochastic neighbor embedding in B1–B4, with the features delineated on the top of each graph. Each color represents a unique diagnosis in the WHO classification scheme. (A1 and B1) Dimensionality reduction with clinical features alone demonstrates only a few visible clusters. (A2 and B2) Incorporating clinical history with histology generates the commencement of clear layering in PCA and discrete clusters with t‐SNE. (A3 and B3) Inclusion of immunohistochemical data improves the capacity of discerning clusters. (A4 and B4) The additional molecular features does not significantly improve the clustering. (C) Individual information gains with all clinical, histologic, immunohistochemical, and molecular features in the glioma simulation model. Age, site, and Ki67 are the features that have the most amount of necessary information for the diagnosis. (D) Information gains with clinical, histologic, immunohistochemical, and molecular information. Histology provides most of the necessary information for the diagnosis. Combining the clinical data with histology and immunohistochemistry provides more than 95% of the necessary information, whereas adding molecular data provides minimal information gain. (E) IDH1/2‐mutated tumors were subsetted, and the mutual information of the features on the X‐axis was quantified as % information on the Y‐axis. The conditional information function call was called on data the 1p19q status.
登录
查看更多内容
Olig2在小儿星形细胞和室室肿瘤中差异表达。
影响因子：
3.9
作者：
Otero, Jose Javier;Rowitch, David;Vandenberg, Scott
通讯作者：
Vandenberg, Scott
影响因子：
--
作者：
SHANNON, CE
通讯作者：
SHANNON, CE
影响因子：
5.8
作者：
Arganda-Carreras, Ignacio;Kaynig, Verena;Seung, H. Sebastian
通讯作者：
Seung, H. Sebastian
影响因子：
3.5
作者：
Torchia, Michael T.;Austin, Daniel C.;Moschetti, Wayne E.
通讯作者：
Moschetti, Wayne E.
影响因子：
3.9
作者：
Fauzi, Mohammad Faizal Ahmad;Gokozan, Hamza Numan;Gurcan, Metin N.
通讯作者：
Gurcan, Metin N.