Deep representation learning of tissue metabolome and computed tomography annotates NSCLC classification and prognosis.

Deep representation learning of tissue metabolome and computed tomography annotates NSCLC classification and prognosis.
复制标题

DOI:
10.1038/s41698-024-00502-3
复制
发表时间:
2024-02-03
影响因子:
7.9
通讯作者:
--
中科院分区:
医学1区
文献类型:
--
作者:

文献摘要

参考文献

相似文献

来自组织代谢组学的丰富化学信息为在细胞和肿瘤微环境水平上阐述组织生理学或肿瘤特征提供了强大的手段。然而,获取此类信息的过程需要侵入性活检,成本高昂,并且可能会延迟临床患者管理。相反,计算机断层扫描 (CT) 是一种临床护理标准,但不能直观地提供组织学或预后信息。此外,将代谢组信息嵌入到 CT 中以随后使用学习到的表示进行分类或预后的能力尚未得到描述。这项研究开发了一个基于深度学习的框架——组织代谢组学放射组学 CT (TMR-CT),通过结合 48 个配对 CT 图像和肿瘤/正常组织代谢物强度来生成 10 个图像嵌入,以仅从 CT 推断代谢物衍生的表示。在临床 NSCLC 环境中,我们确定 TMR-CT 是否会产生增强的特征生成模型,解决未见过的 742 名患者的国际 CT 数据集中的组织学分类/预后任务。 TMR-CT 非侵入性地确定组织学类别 - 腺癌/鳞状细胞癌,F1 分数 = 0.78,并进一步断言患者的预后,c 指数 = 0.72,超越了放射组学模型和单模态 CT 特征提取的深度学习的性能。此外,我们的工作显示了生成信息丰富的生物学启发的 CT 主导特征的潜力,以探索难以获得的组织代谢特征与常规病变衍生图像数据之间的联系。
The rich chemical information from tissue metabolomics provides a powerful means to elaborate tissue physiology or tumor characteristics at cellular and tumor microenvironment levels. However, the process of obtaining such information requires invasive biopsies, is costly, and can delay clinical patient management. Conversely, computed tomography (CT) is a clinical standard of care but does not intuitively harbor histological or prognostic information. Furthermore, the ability to embed metabolome information into CT to subsequently use the learned representation for classification or prognosis has yet to be described. This study develops a deep learning-based framework -- tissue-metabolomic-radiomic-CT (TMR-CT) by combining 48 paired CT images and tumor/normal tissue metabolite intensities to generate ten image embeddings to infer metabolite-derived representation from CT alone. In clinical NSCLC settings, we ascertain whether TMR-CT results in an enhanced feature generation model solving histology classification/prognosis tasks in an unseen international CT dataset of 742 patients. TMR-CT non-invasively determines histological classes - adenocarcinoma/squamous cell carcinoma with an F1-score = 0.78 and further asserts patients’ prognosis with a c-index = 0.72, surpassing the performance of radiomics models and deep learning on single modality CT feature extraction. Additionally, our work shows the potential to generate informative biology-inspired CT-led features to explore connections between hard-to-obtain tissue metabolic profiles and routine lesion-derived image data.
DOI: 10.1016/j.tibs.2014.06.005
发表时间: 2014-08
影响因子: 13.8
作者:
Patra, Krushna C.;Hay, Nissim
通讯作者: Hay, Nissim
基于放射素学的机器学习模型,用于预测稀有癌症的整体和无进展生存:原发性中枢神经系统淋巴瘤患者的案例研究。
DOI: 10.3390/bioengineering10030285
发表时间: 2023-02-22
期刊: Bioengineering (Basel, Switzerland)
影响因子: --
作者:
通讯作者: --
DOI: 10.3390/jpm11070602
发表时间: 2021-06-25
影响因子: --
作者:
Frix AN;Cousin F;Refaee T;Bottari F;Vaidyanathan A;Desir C;Vos W;Walsh S;Occhipinti M;Lovinfosse P;Leijenaar RTH;Hustinx R;Meunier P;Louis R;Lambin P;Guiot J
通讯作者: Guiot J
DOI: 10.1016/j.ebiom.2022.104344
发表时间: 2022-12
期刊: EBIOMEDICINE
影响因子: 11.1
作者:
Hunter, Benjamin;Chen, Mitchell;Ratnakumar, Prashanthi;Alemu, Esubalew;Logan, Andrew;Linton-Reid, Kristofer;Tong, Daniel;Senthivel, Nishanthi;Bhamani, Amyn;Bloch, Susannah;Kemp, Samuel, V;Boddy, Laura;Jain, Sejal;Gareeboo, Shafick;Rawal, Bhavin;Doran, Simon;Navani, Neal;Nair, Arjun;Bunce, Catey;Kaye, Stan;Blackledge, Matthew;Aboagye, Eric O.;Devaraj, Anand;Lee, Richard W.
通讯作者: Lee, Richard W.
DOI: 10.3390/jpm12040601
发表时间: 2022-04-08
影响因子: --
作者:
通讯作者: --