qMTNet+, an Improved qMTNet with Residual Connection for Accelerated Quantitative Magnetization Transfer Imaging
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Purpose: To develop qMTNet+, an improved version of a recently proposed neural network called qMTNet, to accelerate quantitative magnetization transfer (qMT) imaging acquisition and processing. Materials and Methods: Conventional and inter-slice qMT data were acquired with two flip angles at six offset frequencies from seven subjects for developing networks and from four young and four older subjects for testing its generalizability. qMTNet+ was designed to incorporate residual and multi-task learning to improve its performance. qMTNet+ is composed of multiple fully-connected layers. It can simultaneously generate missing MT-weighted images and qMT parameters from undersampled MT images. The network was trained and validated with 7-fold cross-validation. Additional testing with unseen data was performed to assess the generalizability of the network. Performance of qMTNet+ was compared with that of qMTNet-1 and qMTNet-acq for fitting and MT images generation, respectively. Results: qMTNet+ achieved quantitative results that were better than qMTNet across all metrics (peak signal-to-noise ratio, structural similarity index, normalized mean squared error) for both conventional and inter-slice MT data. Produced offset images were better quantitatively than those produced by qMTNet-acq. Conclusion: qMTNet+ improves qMTNet, generating qMT parameters from undersampled MT data with higher agreement with ground truth values. Additionally, qMTNet+ can produce both qMT parameters and unsampled MT images with a single network in an end-to-end manner, obviating the need for separate networks required for qMTNet. qMTNet+ has the potential to accelerate qMT imaging for diagnostic and research purposes.
Abstract
Purpose: To develop qMTNet+, an improved version of a recently proposed neural network called qMTNet, to accelerate quantitative magnetization transfer (qMT) imaging acquisition and processing. Materials and Methods: Conventional and inter-slice qMT data were acquired with two flip angles at six offset frequencies from seven subjects for developing networks and from four young and four older subjects for testing its generalizability. qMTNet+ was designed to incorporate residual and multi-task learning to improve its performance. qMTNet+ is composed of multiple fully-connected layers. It can simultaneously generate missing MT-weighted images and qMT parameters from undersampled MT images. The network was trained and validated with 7-fold cross-validation. Additional testing with unseen data was performed to assess the generalizability of the network. Performance of qMTNet+ was compared with that of qMTNet-1 and qMTNet-acq for fitting and MT images generation, respectively. Results: qMTNet+ achieved quantitative results that were better than qMTNet across all metrics (peak signal-to-noise ratio, structural similarity index, normalized mean squared error) for both conventional and inter-slice MT data. Produced offset images were better quantitatively than those produced by qMTNet-acq. Conclusion: qMTNet+ improves qMTNet, generating qMT parameters from undersampled MT data with higher agreement with ground truth values. Additionally, qMTNet+ can produce both qMT parameters and unsampled MT images with a single network in an end-to-end manner, obviating the need for separate networks required for qMTNet. qMTNet+ has the potential to accelerate qMT imaging for diagnostic and research purposes.
참고문헌 (38)
- Henkelman RM, "Quantitative interpretation of magnetization transfer", Magn Reson Med, 29, pp.759~766 (1993)
- Schmierer K, "Quantitative magnetization transfer imaging in postmortem multiple sclerosis brain", J Magn Reson Imaging, 26, pp.41~51 (2007)
- Levesque IR, "Quantitative magnetization transfer and myelin water imaging of the evolution of acute multiple sclerosis lesions", Magn Reson Med, 63, pp.633~640 (2010)
- Garcia M, "Imaging of primary brain tumors and metastases with fast quantitative 3-dimensional magnetization transfer", J Neuroimaging, 25, pp.1007~1014 (2015)
- Harrison NA, "Quantitative magnetization transfer imaging as a biomarker for effects of systemic inflammation on the brain", Biol Psychiatry, 78, pp.49~57 (2015)
- Cabana J-F, "Quantitative magnetization transfer imaging made easy with qMTLab:software for data simulation, analysis, and visualization", Concepts Magn Reson, 44A, pp.263~277 (2015)
- McLean MA, "Accelerated quantitative magnetization transfer (qMT) imaging", Master, pp.2018
- Gochberg DF, "Quantitative imaging of magnetization transfer using an inversion recovery sequence", Magn Reson Med, 49, pp.501~505 (2003)
- Gochberg DF, "Quantitative magnetization transfer imaging via selective inversion recovery with short repetition times", Magn Reson Med, 57, pp.437~441 (2007)
- Li K, "Optimized inversion recovery sequences for quantitative T1 and magnetization transfer imaging", Magn Reson Med, 64, pp.491~500 (2010)
- Dortch RD, "Quantitative magnetization transfer imaging in human brain at 3T via selective inversion recovery", Magn Reson Med, 66, pp.1346~1352 (2011)
- Kim JW, "Rapid framework for quantitative magnetization transfer imaging with interslice magnetization transfer and dictionary-driven fitting approaches", Magn Reson Med, 82, pp.1671~1683 (2019)
- Barker JW, "Investigation of inter-slice magnetization transfer effects as a new method for MTR imaging of the human brain", PLoS One, 10, pp.e0117101~ (2015)
- Han PK, "Inter-slice blood flow and magnetization transfer effects as a new simultaneous imaging strategy", PLoS One, 10, pp.e0140560~ (2015)
- Park SH, "Alternate ascending", descending directional navigation approach for imaging magnetization transfer asymmetry, 2011, pp.65, 1702~1710
- Park SH, "Brain MR perfusion-weighted imaging with alternate ascending", descending directional navigation, 2011, pp.65, 1578~1591
- 박혜리, "Evaluation of Tumor Blood Flow Using Alternate Ascending", Descending Directional Navigation in Primary Brain Tumors: A Comparison Study with Dynamic Susceptibility Contrast Magnetic Resonance Imaging, 2019, pp.20(2), 275~282[DOI]
- Kim KH, "Feasibility of quantifying arterial cerebral blood volume using multiphase alternate ascending", descending directional navigation(ALADDIN), 2016, pp.11, e0156687~
- Park SH, "Suppression of effects of gradient imperfections on imaging with alternate ascending", descending directional navigation, 2012, pp.68, 1600~1606
- Luu HM, "qMTNet : accelerated quantitative magnetization transfer imaging with artificial neural networks", Magn Reson Med, 85, pp.298~308 (2020)
- Cercignani M, "Optimal acquisition schemes for in vivo quantitative magnetization transfer MRI", Magn Reson Med, 56, pp.803~810 (2006)
- Abadi M, "TensorFlow: a system for large-scale machine learning", Proc the OSDI’16: 12th USENIX Symposium on Operating Systems Design and Implementation, pp.265~283 (2016), USENIX Association
- He K, "Deep residual learning for image recognition", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.770~778 (2016)
- Srivastava N, "Dropout : a simple way to prevent neural networks from overfitting", J Mach Learn Res, 15, pp.1929~1958 (2014)
- Kingma DP, "Adam: a method for stochastic optimization", Proc 3rd International Conference on Learning Representations (ICLR) (2015)
- Wilcoxon F, "Individual comparisons by ranking methods", Biometrics, 1, pp.80~83 (1945)
- Caruana R, "Multitask learning", Machine Learning, 28, pp.41~75 (1997)
- Cohen O, "MR fingerprinting deep reconstruction network(DRONE)", Magn Reson Med, 80, pp.885~894 (2018)
- Yoon J, "Quantitative susceptibility mapping using deep neural network : QSMnet", Neuroimage, 179, pp.199~206 (2018)
- Lee J, "Artificial neural network for myelin water imaging", Magn Reson Med, 83, pp.1875~1883 (2020)
- Heule R, "Multi-parametric artificial neural network fitting of phase-cycled balanced steady-state free precession data", Magn Reson Med, 84, pp.2981~2993 (2020)
- Karras T, "Analyzing and improving the image quality of StyleGAN", Proc IEEE, 2020, 8110~8119
- Chen LC, "DeepLab: semantic image segmentation with deep convolutional nets, atrous convolution, and fully connected CRFs", IEEE Trans Pattern Anal Mach Intell, 40, pp.834~848 (2018)
- Johnson J, "Perceptual losses for realtime style transfer and super-resolution", Proc European Conference on Computer Vision, pp.694~711 (2016)
- Lee D, "Deep residual learning for compressed sensing MRI", IEEE 14th International Symposium on Biomedical Imaging (ISBI 2017), pp.15~18 (2017)
- Lee D, "Deep residual learning for accelerated MRI using magnitude and phase networks", IEEE Trans Biomed Eng, 65, pp.1985~1995 (2018)
- Goodfellow I, "Generative adversarial nets", Advances in Neural Information Processing Systems, pp.2672~2680 (2014)
- Jung W, "Exploring linearity of deep neural network trained QSM: QSMnet+", arXiv preprint arXiv:1909.07716, ~ (2019)