Shou Qinyang, Cen Steven, Chen Nan-Kuei, Ringman John M, Wen Junhao, Kim Hosung, Wang Danny Jj
Laboratory of Functional MRI Technology (LOFT), Stevens Neuroimaging and Informatics Institute, University of Southern California, Los Angeles, CA, United States.
Department of Neurology, Keck School of Medicine, University of Southern California, Los Angeles, CA, United States.
medRxiv. 2024 Jul 3:2024.07.01.24309791. doi: 10.1101/2024.07.01.24309791.
Cerebral blood flow (CBF) measured by arterial spin labeling (ASL) is a promising biomarker for Alzheimer's Disease (AD). ASL data from multiple vendors were included in the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset. However, the M0 images were missing in Siemens ASL data, prohibiting CBF quantification. Here, we utilized a generative diffusion model to impute the missing M0 and validated generated CBF data with acquired data from GE.
A conditional latent diffusion model was trained to generate the M0 image and validate it on an in-house dataset (=55) based on image similarity metrics, accuracy of CBF quantification, and consistency with the physical model. This model was then applied to the ADNI dataset (Siemens: =211) to impute the missing M0 for CBF calculation. We further compared the imputed data (Siemens) and acquired data (GE) regarding regional CBF differences by AD stages, their classification accuracy for AD prediction, and CBF trajectory slopes estimated by a mixed effect model.
The trained diffusion model generated the M0 image with high fidelity (Structural similarity index, SSIM=0.924±0.019; peak signal-to-noise ratio, PSNR=33.348±1.831) and caused minimal bias in CBF values (mean difference in whole brain is 1.07±2.12ml/100g/min). Both generated and acquired CBF data showed similar differentiation patterns by AD stages, similar classification performance, and decreasing slopes with AD progression in specific AD-related regions. Generated CBF data also improved accuracy in classifying AD stages compared to qualitative perfusion data.
INTERPRETATION/CONCLUSION: This study shows the potential of diffusion models for imputing missing modalities for large-scale studies of CBF variation with AD.
通过动脉自旋标记（ASL）测量的脑血流量（CBF）是阿尔茨海默病（AD）一种很有前景的生物标志物。来自多个供应商的ASL数据被纳入阿尔茨海默病神经影像倡议（ADNI）数据集。然而，西门子ASL数据中缺少M0图像，这使得无法进行CBF定量分析。在此，我们利用生成扩散模型来估算缺失的M0，并使用通用电气公司采集的数据对生成的CBF数据进行验证。
训练一个条件潜在扩散模型来生成M0图像，并基于图像相似性指标、CBF定量分析的准确性以及与物理模型的一致性，在一个内部数据集（n = 55）上对其进行验证。然后将该模型应用于ADNI数据集（西门子：n = 211），以估算缺失的M0用于CBF计算。我们进一步比较了估算数据（西门子）和采集数据（通用电气）在AD分期方面的区域CBF差异、它们对AD预测的分类准确性以及通过混合效应模型估计的CBF轨迹斜率。
训练后的扩散模型生成的M0图像具有高保真度（结构相似性指数，SSIM = 0.924±0.019；峰值信噪比，PSNR = 33.348±1.831），并且在CBF值上产生的偏差最小（全脑平均差异为1.07±2.12ml/100g/min）。生成的和采集的CBF数据在AD分期方面均显示出相似的分化模式、相似的分类性能，并且在特定的AD相关区域中随着AD进展斜率下降。与定性灌注数据相比，生成的CBF数据在AD分期分类方面也提高了准确性。
解读/结论：本研究显示了扩散模型在为CBF随AD变化的大规模研究估算缺失模态方面所具有的潜力。
