Original contributionArtificial neural network for suppression of banding artifacts in balanced steady-state free precession MRI
Introduction
Balanced steady state free precession (bSSFP) sequence has been gaining importance in clinical practices due to its high speed, high signal-to-noise ratio (SNR), and minimal distortion. However, bSSFP often suffers from well-known banding artifacts, which are related to the field strength and repetition time (TR). Good shimming is also essential to reduce banding artifacts. Unfortunately, even with optimized scan parameters and good shimming, banding artifacts are frequently unavoidable.
In bSSFP, sum of gradients along each of the three directions (i.e., readout, phase encoding, slice selection) during one TR equals zero [1], [2], indicating no phase contributions of the gradients. The net phase of a proton spin after each TR is therefore affected only by (i) phase evolution per TR due to magnetic field inhomogeneity and (ii) increment in transmission RF phase per TR called “phase cycling” (PC) angle. Typically the former is minimized by using a minimum possible TR and the best shimming condition, while the latter is set to 180° for the residual transverse magnetization from one RF excitation to be used for the following RF excitation in a balanced way. When sum of the two phase terms corresponds to 0° (or integer multiple of 360°) rather than 180°, offset of the RF pulses relative to the precession angle is zero, leading to banding artifacts. Banding artifacts can spatially shift with PC angle, which can be used as a method to avoid banding artifacts in a specific region of interest [3], [4], [5], [6].
Various methods have been introduced to suppress banding artifacts in bSSFP [7], [8], [9], [10], [11], [12], [13], [14], [15], [16]. The most common way is to use multiple bSSFP acquisitions at various PC angles and combine them in a special way, including maximum-intensity projection (MIP) [7], complex sum [8], sum-of-square (SOS) [9], nonlinear averaging [11], sum of free induction decay and primary echo components from Fourier analysis [15], geometric solution [13], linearization with Gauss-Newton algorithm [12], and geometric-algebraic solution [14]. However, banding artifacts cannot be suppressed completely by these methods, especially when number of multiple PC bSSFP datasets is small.
In this study, we proposed a learning-based method to combine the multiple PC bSSFP datasets for better banding artifact suppression. Multilayer perceptron (MLP) is a feedforward artificial neural network consisting of three layers of input, hidden, and output layers. MLP models were trained by input bSSFP datasets acquired from human brain and knee at 3 T, which were separately performed for two and four PC angles. Banding-free bSSFP images were generated by maximum-intensity projection (MIP) of 8 or 12 phase-cycled datasets and were used as targets for training the output layer. The MLP models were applied to another brain and knee datasets acquired with different scan parameters and also to multiple phase-cycled bSSFP functional MRI datasets acquired on rat brain at 9.4 T, in comparison with the conventional MIP method, which has been widely used [11].
Section snippets
Theory
MLP is a variant of artificial neural network, which consists of input, hidden, and output layers [17]. At each layer, there are one or more neurons, and they are fully connected with neurons of neighboring layers. Each connection between neurons is weighted by different weighting factors, and these weights and a nonlinear activation process govern the functional mapping between the sets of inputs and outputs [18].
Fig. 1 shows a schematic diagram of MLP. In order to train MLP, bSSFP datasets
Results
Fig. 3 shows the results of banding artifact correction using MLP. The three methods of MIP, SOS, and MLP suppressed banding artifacts in the simulated ‘bi-tissue’ data. However, MIP and SOS demonstrated visible signal fluctuations in the combined images, especially when the number of PC angles was small (Fig. 3a–b). The signal fluctuations were mostly eliminated in the results of MLP, irrespective of tissue T1/T2 (Fig. 3a–b). MSE values of MLP (1.5 × 10− 5) were much smaller than those of the
Discussion
To our knowledge, it is the first time for an algorithm of artificial neural network to be applied to reducing banding artifacts in bSSFP. Multiple PC acquisitions were used to train the MLP models. The trained MLP models were validated in new bSSFP images acquired with different scan parameters from different body organs, different species, and different field strength and were demonstrated to be superior to existing methods in terms of banding artifact suppression and SNR efficiency. The
Conclusions
Combination of multiple PC datasets with MLP successfully suppressed banding artifacts in bSSFP MRI, while maintaining high SNR and functional signal changes. The MLP models trained with a couple of 3D multiple PC datasets from two different organs in human worked well on datasets acquired with different scan parameters from different subjects, and also those acquired from a different species at different field strength. MLP provided better banding artifact suppression and higher SNR efficiency
Acknowledgements
This work was supported by the Korea Health Technology R&D Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare of South Korea (HI16C1111).
References (25)
- et al.
Balanced steady state free precession for arterial spin labeling MRI: initial experience for blood flow mapping in human brain, retina, and kidney
Magn Reson Imaging
(2013) - et al.
Sensitivity and specificity of high-resolution balanced steady-state free precession fMRI at high field of 9.4 T
Neuroimage
(2011) - et al.
Nonlinear averaging reconstruction method for phase-cycle SSFP
Magn Reson Imaging
(2007) Synthesis methods of multiple phase-cycled SSFP images to reduce the band artifact and noise more reliably
Magn Reson Imaging
(2010)- et al.
Artificial neural networks (the multilayer perceptron) - a review of applications in the atmospheric sciences
Atmos Environ
(1998) - et al.
Multilayer feedforward networks are universal approximators
Neural Netw
(1989) - et al.
MRI breast cancer diagnosis hybrid approach using adaptive ant-based segmentation and multilayer perceptron neural networks classifier
Appl Soft Comput
(2014) - et al.
A hybrid method for MRI brain image classification
Expert Systems with Applications
(2011) - et al.
Principles and applications of balanced SSFP techniques
Eur Radiol
(2003) - et al.
Physiological and functional magnetic resonance imaging using balanced steady-state free precession
Korean J Radiol
(2015)