Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI

Abstract

Purpose To develop new artificial neural networks (ANNs) to accelerate slice encoding for metal artifact correction (SEMAC) MRI. Methods Eight titanium phantoms and 77 patients after brain tumor surgery involving metallic neuro‐plating instruments were scanned using SEMAC at a 3T Skyra scanner. For the phantoms, proton‐density, T1‐, and T2‐weighted images were acquired for developing both multilayer perceptron (MLP) and convolutional neural network (CNN). For the patients, T2‐weighted images were acquired for developing CNN. All networks were trained with the SEMAC factor 4 or 6 as input and the factor 12 as label, yielding an acceleration factor of 3 or 2. Performance of the CNN model was compared against parallel imaging and compressed sensing on the phantom datasets. Two extra T1‐weighted in vivo sets were acquired to investigate generalizability of the models to different contrasts. Results Both multilayer perceptron and CNN provided artifact‐suppressed images better than the input images and comparable to the label images visually and quantitatively, a trend observable regardless of input SEMAC factor and image type (P < .01). CNN suppressed the artifacts better than multilayer perceptron, parallel imaging, and compressed sensing (P < .01). Tests on the patient datasets demonstrated clear metal artifact suppression visually and quantitatively (P < .01). Tests on T1 datasets also demonstrated clear visual metal artifact suppression. Conclusion Our study introduced a new effective way of artificial neural networks to accelerate SEMAC MRI while maintaining the comparable quality of metal artifact suppression. Application on the preliminary patient datasets proved the feasibility in clinical usage, which warrants further investigation.

[Retracted] Analysis of Sports Injury Estimation Model Based on Mutation Fuzzy Neural Network

“…With the development of science and technology, it has been widely used in various fields of life [ 1 – 3 ]. The algorithm models of these neural networks are called artificial neural network model, or neural network model [ 4 , 5 ]. In many neural network models, the development of fuzzy neural network has been concerned [ 6 , 7 ].…”

Design and Preliminary Realization of a Screening and Early Warning Health Management System for Populations at High Risk for Depression

“…It can automatically find patterns according to the input data, so as to make reliable predictions. Multilayer Perceptron (MLP) [ 20 , 21 ] is a powerful extension of perceptron and has the ability to approximate the nonlinear relationship between input layer and output layer with arbitrary accuracy. MLP [ 22 ] is a feedforward supervised structure, as shown in Figure 2 .…”

Review: Noise and artifact reduction for MRI using deep learning

“…The DL-based approach is also useful in solving other complicated problems such as removing Gibbs [44][45][46][47] , metal [48][49][50] , and banding artifact 51 . Zhang et al proposed the CNN-based Gibbs artifact, which can be an alternative to conventional k-space domain filtering methods 45 .…”