Deep flow-net for EPI distortion estimation
14
Citation
27
Reference
10
Related Paper
Citation Trend
Abstract:
Geometric distortions along the phase encoding direction caused by off-resonant spins are a major issue in EPI based functional and diffusion imaging. The widely used blip up/down approach estimates the underlying distortion field from a pair of images with inverted phase encoding direction. Typically, iterative methods are used to find a solution to the ill-posed problem of finding the displacement field that maps up/down acquisitions onto each other. Here, we explore the use of a deep convolutional network to estimate the displacement map from a pair of input images.We trained a deep convolutional U-net architecture that was previously used to estimate optic flow between moving images to learn to predict the distortion map from an input pair of distorted EPI acquisitions. During the training step, the network minimizes a loss function (similarity metric) that is calculated from corrected input image pairs. This approach does not require the explicit knowledge of the ground truth distortion map, which is difficult to get for real life data.We used data from a total of Ntrain = 22 healthy subjects to train our network. A separate dataset of Ntest = 12 patients including some with abnormal findings and unseen acquisition modes, e.g. LR-encoding, coronal orientation) was reserved for testing and evaluation purposes. We compared our results to FSL's topup function with default parameters that served as the gold standard. We found that our approach results in a correction accuracy that is virtually identical to the optimum found by an iterative search, but with reduced computational time.By using a deep convolutional network, we can reduce the processing time to a few seconds per volume, which is significantly faster than iterative approaches like FSL's topup which takes around 10min on the same machine (but using only 1 CPU). This facilitates the use of a blip up/down scheme for all diffusion-weighted acquisitions and potential real-time EPI distortion correction without sacrificing accuracy.Keywords:
Net (polyhedron)
Distortion (music)
As you know, the .NET platform supports two types of .NET assemblies, static and dynamic. This chapter and the next are about working with dynamic .NET assemblies; it covers the purpose and benefits of dynamic .NET assemblies and how to perform fundamental tasks with them. Also, you will learn how to define a dynamic .NET assembly, a dynamic .NET module, a dynamic .NET reference type, and a dynamic .NET field member.
Net (polyhedron)
Dynamic simulation
Cite
Citations (0)
Net (polyhedron)
Social Connectedness
Limit set
Cite
Citations (8)
Net (polyhedron)
Safety net
Net gain
Cite
Citations (0)
Net (polyhedron)
Morphism
Cite
Citations (0)
Net (polyhedron)
Net gain
Safety net
Cite
Citations (0)
Net (polyhedron)
Square (algebra)
Cite
Citations (0)
본 논문에서는 원문 폰트를 특정한 아날로그 폰트 스타일로 변환하는 타이포그래피 변환 문제에 대해 연구한다. 타이포그래피 변환 문제를 해결하기 위해 이 문제를 이미지와 이미지 번역 문제로 치환하고 GAN을 기반으로 한 언밸런스 형 u-net 아키텍처를 제안한다. 기존의 밸런스 형 u-net과는 달리 제안하는 아키텍처는 언밸런스 형 u-net을 포함한 두 개의 서브넷으로 구성된다. (1)언밸런스 형 u-net은의미 및 구조 정보를 유지하면서 특정 글꼴 스타일을 다른 스타일로 변환한다. (2) GAN은 L1 손실, 상수손실 및 원하는 목표 글꼴을 생성하는 데 도움이 되는 이진 GAN 손실을 포함하는 복합 손실 함수를 사용한다. 실험결과 제안하는 모델인 언밸런스 형 u-net이 밸런스 형 u-net 보다 cheat loss에서 빠른 수렴속도와 안정적인 트레이닝 손실을 보였고 generate loss에서 트레이닝 손실을 안정적으로 줄여서 모델 성능 하락 문제를 해결하였다.
Net (polyhedron)
.NET Framework
Safety net
Net worth
Cite
Citations (2)
Net (polyhedron)
.NET Framework
Net worth
Cite
Citations (24)
Net (polyhedron)
Safety net
Square (algebra)
Table (database)
Cite
Citations (0)
The .NET Core platform supports two types of .NET assemblies, static and dynamic. This chapter covers how to work with the static .NET assembly, the purpose and benefits of the static .NET assembly, the structural organization of a .NET assembly's metadata, and how to do fundamental tasks with .NET assemblies.
Net (polyhedron)
Cite
Citations (0)