Motion artifact correction for MR images based on convolutional neural network

Abstract

Magnetic resonance imaging (MRI) is a common way to diagnose related diseases. However, the magnetic resonance (MR) images are easily defected by motion artifacts in their acquisition process, which affects the clinicians’ diagnosis. In order to correct the motion artifacts of MR images, we propose a convolutional neural network (CNN)-based method to solve the problem. Our method achieves a mean peak signal-to-noise ratio (PSNR) of (35.212±3.321) dB and a mean structural similarity (SSIM) of 0.974 ± 0.015 on the test set, which are better than those of the comparison methods.

References

[1] AndreJ B, BresnahanB W, BashaM M, et al.. Toward quantifying the prevalence, severity, and cost associated with patient motion during clinical MR examinations. Journal of the American college of radiology, 2015, 12(7):689-695

[2] GodenschwwgwrF, KagebeinU, StuchtD, et al.. Motion correction in MRI of the brain. Physics in medicine & biology, 2016, 61(5):R32-R56

[3] Van Der KouweA J W, BennerT, DaleA M. Real-time rigid body motion correction and shimming using cloverleaf navigators. Magnetic resonance in medicine: an official journal of the international society for magnetic resonance in medicine, 2006, 56(5):1019-1032

[4] WelchE B, ManducaA, GrimmR C, et al.. Spherical navigator echoes for full 3D rigid body motion measurement in MRI. Magnetic resonance in medicine: an official journal of the international society for magnetic resonance in medicine, 2002, 47(1):32-41

[5] LYU Q, SHAN H, XIE Y, et al. Cine cardiac MRI motion artifact reduction using a recurrent neural network[EB/OL]. (2020-06-23)[2021-09-20]. https://arxiv.org/abs/2006.12700v1.

[6] TamadaD, KromreyM L, IchikawaS, et al.. Motion artifact reduction using a convolutional neural network for dynamic contrast enhanced MR imaging of the liver. Magnetic resonance in medical sciences, 2020, 19(1):64-76

[7] ZhangK, ZuoW, ChenY, et al.. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoising. IEEE transactions on image processing, 2017, 26(7):3142-3155

[8] PAWAR K, CHEN Z, SHAH N J, et al. MoCoNet: motion correction in 3D MPRAGE images using a convolutional neural network approach[EB/OL]. (2018-07-29) [2021-09-20]. https://arxiv.org/abs/1807.10831v1.

[9] LiuJ, KocakM, SupanichM, et al.. Motion artifacts reduction in brain MRI by means of a deep residual network with densely connected multi-resolution blocks (DRN-DCMB). Magnetic resonance imaging, 2020, 71: 69-79

[10] RonnebergerO, FischerP, BroxT. U-net: convolutional networks for biomedical image segmentation, 2015, Berlin, Heidelberg, Springer-Verlag: 234-241

[11] WooS, ParkJ, LeeJ Y, et al.. CBAM: convolutional block attention module, 2018, Berlin, Heidelberg, Springer-Verlag: 3-19

[12] HeK, ZhangX, RenS, et al.. Deep residual learning for image recognition, 2016, New York, IEEE: 770-778

[13] HeK, ZhangX, RenS, et al.. Delving deep into rectifiers: surpassing human-level performance on imagenet classification, 2015, New York, IEEE: 1026-1034

[14] KINGMA D P, BA J. Adam: a method for stochastic optimization[EB/OL]. (2017-06-30)[2021-09-20]. https://arxiv.org/pdf/1412.6980.pdf.

[15] ZhaoH, GalloO, FrosioI, et al.. Loss functions for image restoration with neural networks. IEEE transactions on computational imaging, 2016, 3(1):47-57

[16] LongJ, ShelhamerE, DarrellT. Fully convolutional networks for semantic segmentation, 2015, New York, IEEE: 3431-3440