Abstract
Today, many MRI reconstruction techniques exist for undersampled MRI data. Regularization-based techniques inspired by compressed sensing allow for the reconstruction of undersampled data that would lead to an ill-posed reconstruction problem. Parallel imaging enables the reconstruction of MRI images from undersampled multi-coil data that leads to a well-posed reconstruction problem. Autocalibrating pMRI techniques encompass pMRI techniques where no explicit knowledge of the coil sensivities is required. A first purpose of this paper is to derive a novel autocalibration approach for pMRI that allows for the estimation and use of smooth, but high-bandwidth coil profiles instead of a compactly supported kernel. These high-bandwidth models adhere more accurately to the physics of an antenna system. The second purpose of this paper is to demonstrate the feasibility of a parameter-free reconstruction algorithm that combines autocalibrating pMRI and compressed sensing. Therefore, we present several techniques for automatic parameter estimation in MRI reconstruction. Experiments show that a higher reconstruction accuracy can be had using high-bandwidth coil models and that the automatic parameter choices yield an acceptable result.
Highlights
In this paper, a novel MRI reconstruction algorithm is presented
Automatic MRI reconstruction The first goal of this paper is to present a single reconstruction technique that tackles a very wide scope of typical reconstruction problems jointly and automatically: Problems associated with advanced MRI reconstruction are sub-Nyquist sampling (Section 1.1), non-uniform sampling (Section 1.2), noise (Section 1.3) and parallel imaging (Section 1.4)
There is no real solution to this problem, so we propose the same assumption as in other parallel MR imaging (pMRI) reconstruction techniques [6,7], namely that reconstruction is made through the sum of squares of the coil images so that for a pixel with index p, we sum over the coils i: rffiX ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi bp~
Summary
The current state of the art in MRI reconstruction consists of many excellent algorithms, but these algorithms require manual intervention for one or more parameter settings, which can be a significant downside. Parameters encompass things such as denoising vs datafit strength, calibration region selection, restrictive k-space trajectory input, pMRI autocalibration kernel size, etc. They arise because reconstruction algorithms attempt to tackle important problems that are associated with different types of MRI acquisition and reconstruction. The current solutions to these problems entail respectively compressed sensing reconstruction (including but not limited to [1,2]), regridded reconstruction (including but not limited to [2,3]), (image-domain) noise estimation [4] and different pMRI techniques [5]
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have