使用自适应模型对双梯度回波序列中纵向弛豫率变化（ΔR）进行磁共振成像估计

Magnetic Resonance Imaging Estimation of Longitudinal Relaxation Rate Change (ΔR) in Dual Gradient Echo Sequences Using an Adaptive Model.

作者信息

Bagher-Ebadian H, Nejad-Davarani S P, Ali M M, Brown S, Makki M, Jiang Q, Noll D C, Ewing J R

机构信息

Physics Dept., Oakland University, Rochester, MI 48309 and Dept. Nuclear Engineering, Shiraz University, Shiraz; Siamak P. Nejad-Davarani - Dept. Neurology, Henry Ford Hospital and Dept. Biomedical Eng. University of Michigan, Ann Arbor; Meser M. Ali - Dept. Radiology, Henry Ford Hospital, Stephen L. Brown - Dept. Radiation Oncology, Henry Ford Hospital, Malek Makki, - Dept. Diagnostic Imaging, University Children Hospital of Zurich, Zurich, Switzerland, Quan Jiang - Dept. Neurology, Henry Ford Hospital; Douglas. C. Noll - Dept. Biomedical Engineering, University of Michigan, Ann Arbor, James R. Ewing Dept. Neurology, Henry Ford Hospital, Dept. Neurology, Wayne State University., and Dept. of Physics, Oakland University.

出版信息

Proc Int Jt Conf Neural Netw. 2011;2011:2501-2506. doi: 10.1109/IJCNN.2011.6033544.

DOI:10.1109/IJCNN.2011.6033544

PMID:25285243

原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC4181328/

Abstract

Magnetic Resonance Imaging (MRI) estimation of contrast agent concentration in fast pulse sequences such as Dual Gradient Echo (DGE) imaging is challenging. An Adaptive Neural Network (ANN) was trained with a map of contrast agent concentration estimated by Look-Locker (LL) technique (modified version of inversion recovery imaging) as a gold standard. Using a set of features extracted from DGE MRI data, an ANN was trained to create a voxel based estimator of the time trace of CA concentration. The ANN was trained and tested with the DGE and LL information of six Fisher rats using a K-Fold Cross-Validation (KFCV) method with 60 folds and 10500 samples. The Area Under the Receiver Operator Characteristic Curve (AUROC) for 60 folds was used for training, testing and optimization of the ANN. After training and optimization, the optimal ANN (4:7:5:1) produced maps of CA concentration which were highly correlated () with the CA concentration estimated by the LL technique. The estimation made by the ANN had an excellent overall performance (AUROC = 0.870).

摘要

在双梯度回波（DGE）成像等快速脉冲序列中，通过磁共振成像（MRI）估算造影剂浓度具有挑战性。以通过Look-Locker（LL）技术（反转恢复成像的改进版本）估算的造影剂浓度图作为金标准，训练了一个自适应神经网络（ANN）。利用从DGE MRI数据中提取的一组特征，训练ANN以创建基于体素的造影剂浓度时间轨迹估计器。使用具有60折和10500个样本的K折交叉验证（KFCV）方法，对六只Fisher大鼠的DGE和LL信息进行了ANN的训练和测试。用于训练、测试和优化ANN的是60折的受试者操作特征曲线下面积（AUROC）。经过训练和优化后，最优的ANN（4:7:5:1）生成的造影剂浓度图与通过LL技术估算的造影剂浓度高度相关（）。ANN的估计具有出色的整体性能（AUROC = 0.870）。