Deep Learned Diffusion Tensor Imaging

Hongyu Li; Chaoyi Zhang; Zifei Liang; Dong Liang; Bowen Shen; Yulin Ge; Jiangyang Zhang; Ruiying Liu; Peizhou Huang; Sunil Gaire; Xiaoliang Zhang; Leslie Ying

Deep Learned Diffusion Tensor Imaging

Hongyu Li, Chaoyi Zhang, Zifei Liang, Dong Liang, Bowen Shen, Yulin Ge, Jiangyang Zhang, Ruiying Liu, Peizhou Huang, Sunil Gaire, Xiaoliang Zhang, Leslie Ying

Electrical and Computer Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Diffusion tensor imaging typically requires acquisition of a large number of diffusion weighted images (DWI) for accurate fitting of the tensor model due to the issue of low SNR. This abstract presents a deep learning method to generate FA color map showing the primary diffusion directions from very few DWIs. The method uses deep convolutional neural networks to learn the nonlinear relationship between the DWIs and the FA color maps, bypassing the conventional DTI models. Experimental results show that the proposed method is able to generate FA color maps from only 6 DWIs with quality comparable to results from 270 DWIs using conventional tensor fitting.

Original language	English
Title of host publication	Unknown book
State	Published - 2019

Cite this

@inproceedings{7632a6ff2d71493d804dcf13275c202e,

title = "Deep Learned Diffusion Tensor Imaging",

abstract = "Diffusion tensor imaging typically requires acquisition of a large number of diffusion weighted images (DWI) for accurate fitting of the tensor model due to the issue of low SNR. This abstract presents a deep learning method to generate FA color map showing the primary diffusion directions from very few DWIs. The method uses deep convolutional neural networks to learn the nonlinear relationship between the DWIs and the FA color maps, bypassing the conventional DTI models. Experimental results show that the proposed method is able to generate FA color maps from only 6 DWIs with quality comparable to results from 270 DWIs using conventional tensor fitting.",

author = "Hongyu Li and Chaoyi Zhang and Zifei Liang and Dong Liang and Bowen Shen and Yulin Ge and Jiangyang Zhang and Ruiying Liu and Peizhou Huang and Sunil Gaire and Xiaoliang Zhang and Leslie Ying",

year = "2019",

language = "English",

booktitle = "Unknown book",

}

TY - GEN

T1 - Deep Learned Diffusion Tensor Imaging

AU - Li, Hongyu

AU - Zhang, Chaoyi

AU - Liang, Zifei

AU - Liang, Dong

AU - Shen, Bowen

AU - Ge, Yulin

AU - Zhang, Jiangyang

AU - Liu, Ruiying

AU - Huang, Peizhou

AU - Gaire, Sunil

AU - Zhang, Xiaoliang

AU - Ying, Leslie

PY - 2019

Y1 - 2019

N2 - Diffusion tensor imaging typically requires acquisition of a large number of diffusion weighted images (DWI) for accurate fitting of the tensor model due to the issue of low SNR. This abstract presents a deep learning method to generate FA color map showing the primary diffusion directions from very few DWIs. The method uses deep convolutional neural networks to learn the nonlinear relationship between the DWIs and the FA color maps, bypassing the conventional DTI models. Experimental results show that the proposed method is able to generate FA color maps from only 6 DWIs with quality comparable to results from 270 DWIs using conventional tensor fitting.

AB - Diffusion tensor imaging typically requires acquisition of a large number of diffusion weighted images (DWI) for accurate fitting of the tensor model due to the issue of low SNR. This abstract presents a deep learning method to generate FA color map showing the primary diffusion directions from very few DWIs. The method uses deep convolutional neural networks to learn the nonlinear relationship between the DWIs and the FA color maps, bypassing the conventional DTI models. Experimental results show that the proposed method is able to generate FA color maps from only 6 DWIs with quality comparable to results from 270 DWIs using conventional tensor fitting.

M3 - Conference contribution

BT - Unknown book

ER -

Deep Learned Diffusion Tensor Imaging

Abstract

Fingerprint

Cite this