Improving Sensitivity of Arterial Spin Labeling Perfusion MRI in Alzheimer's Disease Using Transfer Learning of Deep Learning-Based ASL Denoising.
Improving Sensitivity of Arterial Spin Labeling Perfusion MRI in Alzheimer's Disease Using Transfer Learning of Deep Learning-Based ASL Denoising.
复制标题
DOI：
10.1002/jmri.27984
复制
发表时间：
2022-06
影响因子：
4.4
通讯作者：
Wang, Ze
中科院分区：
文献类型：
--
作者：
Zhang, Lei;Xie, Danfeng;Li, Yiran;Camargo, Aldo;Song, Donghui;Lu, Tong;Jeudy, Jean;Dreizin, David;Melhem, Elias R.;Wang, Ze
动脉自旋标记灌注MRI （ASL MRI）通过深度学习（DL）去噪往往面临患者训练数据不足的问题。一种解决方案是使用更广泛可用的健康受试者数据来训练深度学习模型，并将其转换为患者数据。为了评估基于dl的ASL MRI去噪方法（DLASL）的可移植性，回顾性分析了来自3个队列的428名受试者（189名女性）。3T基于二维回波平面成像（EPI）的伪连续ASL （PCASL）和基于二维EPI的脉冲ASL （PASL）序列。DLASL使用健康青年PCASL数据（数据集1:250 / 30受试者作为训练/验证集）进行训练，并直接（DTF）转移到正常对照组（NC）和AD组（数据集2:45受试者测试集）的PCASL数据。对NC和AD的PASL数据集3（103受试者测试集）进行微调（DLASLFT）和测试。采用现有的非深度分类方法（NonDL）进行比较。比较NC和AD的ASL MRI脑血流（CBF）图像，以评估AD的特征性低灌注（低CBF）模式。使用峰值t值和超阈值聚类大小检测低灌注的CBF图像质量和CBF图灵敏度是结果度量。采用配对t检验、双样本t检验、单因素方差分析、杜奇诚实显著差异和线性混合效应模型。P<0.05为差异有统计学意义。数据集2的平均噪比（CNR）显示，DTF优于NonDL （AD: 3.38 vs. 2.64, NC: 3.80 vs. 3.36）。在数据集3上，通过平均CNR （AD: 2.45 vs. 1.87, NC: 2.54 vs. 2.17）和平均放射学评分（2.86 vs. 2.44）测量，DLASLFT优于NonDL。两个测试集的图像质量都有显著改善。与NonDL相比，DTF和DLASLFT提高了检测ad相关低灌注模式的灵敏度。我们证明了DLASL在不同的ASL序列和不同的人群之间的可转移性。
Arterial Spin Labeling Perfusion MRI (ASL MRI) denoising through deep learning (DL) often faces insufficient training data from patients. One solution is to train DL models using healthy subjects’ data which are more widely available and transfer them to patients’ data. To evaluate the transferability of a DL-based ASL MRI denoising method (DLASL) Retrospective. 428 subjects (189 females) from 3 cohorts. 3T 2D Echo-planar imaging (EPI)-based pseudo-continuous ASL (PCASL) and 2D EPI-based pulsed ASL (PASL) sequences. DLASL was trained using young healthy adults’ PCASL data (Dataset 1: 250 / 30 subjects as training / validation set) and was directly transferred (DTF) to PCASL data from Dataset 2 (45 subjects test set) of normal controls (NC) and AD groups. DLASL was fine-tuned (DLASLFT) and tested on PASL data from Dataset 3 (103 subjects test set) of NC and AD. An existing non-DL method (NonDL) was used for comparison. Cerebral blood flow (CBF) images from ASL MRI were compared between NC and AD to assess characteristic hypoperfusion (lower CBF) patterns in AD. CBF image quality and CBF map sensitivity for detecting hypoperfusion using peak t-value and suprathreshold cluster size are outcome measures. Paired t-test, two-sample t-test, one-way ANOVA, and Tukey Honestly Significant Difference, and linear mixed-effects models were used. P<0.05 was considered statistically significant. Mean contrast-to-noise-ratio (CNR) of Dataset 2 showed that DTF outperformed NonDL (AD: 3.38 vs. 2.64, NC: 3.80 vs. 3.36). On Dataset 3, DLASLFT outperformed NonDL measured by mean CNR (AD: 2.45 vs. 1.87, NC: 2.54 vs. 2.17) and mean radiologic score (2.86 vs. 2.44). Image quality improvement was significant on both test sets. DTF and DLASLFT improved sensitivity for detecting AD-related hypoperfusion patterns compared with NonDL. We demonstrated the DLASL’s transferability across different ASL sequences and different populations.
登录
查看更多内容
影响因子：
4.8
作者：
Wang, Ze
通讯作者：
Wang, Ze
影响因子：
5.7
作者：
Kim, Byungjai;Schaer, Michael;Heo, Hye-Young
通讯作者：
Heo, Hye-Young
动脉自旋标记血流MRI：其在阿尔茨海默氏病早期表征中的作用。
DOI：
10.3233/jad-2010-091699
发表时间：
2010
期刊：
Journal of Alzheimer's disease : JAD
影响因子：
--
作者：
Alsop DC;Dai W;Grossman M;Detre JA
通讯作者：
Detre JA
影响因子：
5.7
作者：
Fonov V;Evans AC;Botteron K;Almli CR;McKinstry RC;Collins DL;Brain Development Cooperative Group
通讯作者：
Brain Development Cooperative Group
影响因子：
2.5
作者：
Wang, Ze;Aguirre, Geoffrey K.;Detre, John A.
通讯作者：
Detre, John A.