Deep learning super-resolution magnetic resonance spectroscopic imaging of brain metabolism and mutant isocitrate dehydrogenase glioma.
Deep learning super-resolution magnetic resonance spectroscopic imaging of brain metabolism and mutant isocitrate dehydrogenase glioma.
复制标题
DOI：
10.1093/noajnl/vdac071
复制
发表时间：
2022-01
期刊：
影响因子：
--
通讯作者：
Andronesi, Ovidiu C.
中科院分区：
文献类型：
--
作者：
Li, Xianqi;Strasser, Bernhard;Neuberger, Ulf;Vollmuth, Philipp;Bendszus, Martin;Wick, Wolfgang;Dietrich, Jorg;Batchelor, Tracy T.;Cahill, Daniel P.;Andronesi, Ovidiu C.
关键词：
磁共振波谱成像(MRSI)可以用于胶质瘤患者，以绘制与IDH1，2突变相关的代谢变化图，这是胶质瘤诊断的中心标准。这项研究的目的是使用深度学习实现超分辨率(SR)磁共振成像，以成像突变型IDH胶质瘤患者的肿瘤代谢。我们提出了一种基于产生式对抗网络(GAN)的深度学习方法，该方法使用UNT作为生成器网络，将MRSI向上采样4倍。神经网络在75例胶质瘤患者的模拟代谢图像上进行训练。根据20例脑胶质瘤患者和10例健康对照的3T磁共振成像数据，采用优化的全脑3DMRSI方案检测d-2-羟基戊二酸(2HG)，对深层神经元网络的性能进行评估。为了进一步增强代谢图的结构细节，我们使用了来自高分辨率解剖磁共振成像的先前信息。通过峰值信噪比(PSNR)、结构相似性指数(SSIM)、基于特征的相似性指数(FSIM)和平均意见评分(MOS)的Mann-Whitney U检验将SR MRSI与地面真实情况进行比较。与传统的插值法相比，深度学习随机共振的PSNR提高了17%，SSIM提高了5%，FSIM提高了7%，MOS提高了30%。在突变型IDH胶质瘤患者中，所提出的方法提供了2HG图的最高分辨率，以清楚地描绘肿瘤边缘和肿瘤的异质性。我们的结果表明，所提出的深度学习方法在提高代谢产物图谱的空间分辨率方面是有效的。患者结果表明，这可能在影像引导的精确肿瘤学治疗方面具有巨大的临床潜力。
Magnetic resonance spectroscopic imaging (MRSI) can be used in glioma patients to map the metabolic alterations associated with IDH1,2 mutations that are central criteria for glioma diagnosis. The aim of this study was to achieve super-resolution (SR) MRSI using deep learning to image tumor metabolism in patients with mutant IDH glioma. We developed a deep learning method based on generative adversarial network (GAN) using Unet as generator network to upsample MRSI by a factor of 4. Neural networks were trained on simulated metabolic images from 75 glioma patients. The performance of deep neuronal networks was evaluated on MRSI data measured in 20 glioma patients and 10 healthy controls at 3T with a whole-brain 3D MRSI protocol optimized for detection of d-2-hydroxyglutarate (2HG). To further enhance structural details of metabolic maps we used prior information from high-resolution anatomical MR imaging. SR MRSI was compared to ground truth by Mann–Whitney U-test of peak signal-to-noise ratio (PSNR), structure similarity index measure (SSIM), feature-based similarity index measure (FSIM), and mean opinion score (MOS). Deep learning SR improved PSNR by 17%, SSIM by 5%, FSIM by 7%, and MOS by 30% compared to conventional interpolation methods. In mutant IDH glioma patients proposed method provided the highest resolution for 2HG maps to clearly delineate tumor margins and tumor heterogeneity. Our results indicate that proposed deep learning methods are effective in enhancing spatial resolution of metabolite maps. Patient results suggest that this may have great clinical potential for image guided precision oncology therapy.
登录
查看更多内容
影响因子：
10.6
作者：
Mason, Allister;Rioux, James;Beyea, Steven
通讯作者：
Beyea, Steven
影响因子：
64.8
作者：
通讯作者：
--
基于斑块的MR光谱图像的超分辨率：应用于多发性硬化症。
影响因子：
4.3
作者：
Jain S;Sima DM;Sanaei Nezhad F;Hangel G;Bogner W;Williams S;Van Huffel S;Maes F;Smeets D
通讯作者：
Smeets D
影响因子：
16.6
作者：
Andronesi OC;Arrillaga-Romany IC;Ly KI;Bogner W;Ratai EM;Reitz K;Iafrate AJ;Dietrich J;Gerstner ER;Chi AS;Rosen BR;Wen PY;Cahill DP;Batchelor TT
通讯作者：
Batchelor TT
影响因子：
2.9
作者：
Maudsley AA;Andronesi OC;Barker PB;Bizzi A;Bogner W;Henning A;Nelson SJ;Posse S;Shungu DC;Soher BJ
通讯作者：
Soher BJ