Abstract: Diffusion tensor imaging (DTI) is a widely used imaging technique for mapping living human brain tissue’s microstructure and structural connectivity. Recently, deep learning methods have been proposed to rapidly estimate diffusion tensors (DTs) using only a small quantity of diffusion-weighted (DW) images. However, these methods typically use the DW images obtained with fixed q-space sampling schemes as the training data, limiting the application scenarios of such methods. To address this issue, we develop a new deep neural network called q-space-coordinate-guided diffusion tensor imaging (QCG-DTI), which can efficiently and correctly estimate DTs under flexible q-space sampling schemes. First, we propose a q-space-coordinate-embedded feature consistency strategy to ensure the correspondence between q-space-coordinates and their respective DW images. Second, a q-space-coordinate fusion (QCF) module is introduced which efficiently embeds q-space-coordinates into multiscale features of the corresponding DW images by linearly adjusting the feature maps along the channel dimension, thus eliminating the dependence on fixed diffusion sampling schemes. Finally, a multiscale feature residual dense (MRD) module is proposed which enhances the network's feature extraction and image reconstruction capabilities by using dual-branch convolutions with different kernel sizes to extract features at different scales. Compared to state-of-the-art methods that rely on a fixed sampling scheme, the proposed network can obtain high-quality diffusion tensors and derived parameters even using DW images acquired with flexible q-space sampling schemes. Compared to state-of-the-art deep learning methods, QCG-DTI reduces the mean absolute error by approximately 15% on fractional anisotropy and around 25% on mean diffusivity.

Q空间坐标引导的神经网络从最小数量扩散加权图像中实现高保真扩散张量估计

[1]Alexander AL, Lee JE, Lazar M, et al., 2007. Diffusion tensor imaging of the brain. Neurotherapeutics, 4(3):316-329.

[2]Aliotta E, Nourzadeh H, Sanders J, et al., 2019. Highly accelerated, model-free diffusion tensor MRI reconstruction using neural networks. Med Phys, 46(4):1581-1591.

[3]Barnea-Goraly N, Kwon H, Menon V, et al., 2004. White matter structure in autism: preliminary evidence from diffusion tensor imaging. Biol Psychiat, 55(3):323-326.

[4]Barnea-Goraly N, Menon V, Eckert M, et al., 2005. White matter development during childhood and adolescence: a cross-sectional diffusion tensor imaging study. Cereb Cort, 15(12):1848-1854.

[5]Basser PJ, Pajevic S, Pierpaoli C, et al., 2000. In vivo fiber tractography using DT-MRI data. Magn Reson Med, 44(4):625-632.

[6]Bastiani M, Andersson JLR, Cordero-Grande L, et al., 2019. Automated processing pipeline for neonatal diffusion MRI in the developing Human Connectome Project. NeuroImage, 185:750-763.

[7]Caruyer E, Lenglet C, Sapiro G, et al., 2013. Design of multishell sampling schemes with uniform coverage in diffusion MRI. Magn Reson Med, 69(6):1534-1540.

[8]Chen YW, Li JL, Lu QQ, et al., 2025. Spherical harmonics-based deep learning achieves generalized and accurate diffusion tensor imaging. IEEE J Biomed Health Inform, 29(1):456-461.

[9]Dale AM, Fischl B, Sereno MI, 1999. Cortical surface-based analysis: I. segmentation and surface reconstruction. NeuroImage, 9(2):179-194.

[10]de Almeida Martins JP, Nilsson M, Lampinen B, et al., 2021. Neural networks for parameter estimation in microstructural MRI: application to a diffusion-relaxation model of white matter. NeuroImage, 244:118601.

[11]Donald Tournier J, Calamante F, Connelly A, 2012. MRtrix: diffusion tractography in crossing fiber regions. Int J Imag Syst Technol, 22(1):53-66.

[12]Elam JS, Glasser MF, Harms MP, et al., 2021. The Human Connectome Project: a retrospective. NeuroImage, 244:118543.

[13]Eriksson SH, Rugg-Gunn FJ, Symms MR, et al., 2001. Diffusion tensor imaging in patients with epilepsy and malformations of cortical development. Brain, 124(3):617-626.

[14]Fischl B, 2012. FreeSurfer. NeuroImage, 62(2):774-781.

[15]Fischl B, Sereno MI, Dale AM, 1999. Cortical surface-based analysis: II: inflation, flattening, and a surface-based coordinate system. NeuroImage, 9(2):195-207.

[16]Garyfallidis E, Brett M, Amirbekian B, et al., 2014. Dipy, a library for the analysis of diffusion MRI data. Front Neuroinform, 8:8.

[17]Gibbons EK, Hodgson KK, Chaudhari AS, et al., 2019. Simultaneous NODDI and GFA parameter map generation from subsampled q-space imaging using deep learning. Magn Reson Med, 81(4):2399-2411.

[18]Golkov V, Dosovitskiy A, Sperl JI, et al., 2016. q-space deep learning: twelve-fold shorter and model-free diffusion MRI scans. IEEE Trans Med Imag, 35(5):1344-1351.

[19]Ha S, Lyu I, 2022. SPHARM-Net: spherical harmonics- based convolution for cortical parcellation. IEEE Trans Med Imag, 41(10):2739-2751.

[20]Jones DK, 2004. The effect of gradient sampling schemes on measures derived from diffusion tensor MRI: a Monte Carlo study. Magn Reson Med, 51(4):807-815.

[21]Jones DK, 2009. Chapter 3 - Gaussian modeling of the diffusion signal. In: Johansen-Berg H, Behrens TEJ (Eds.), Diffusion MRI: from Quantitative Measurement to in vivo Neuroanatomy. Elsevier, Amsterdam, the Netherlands, p.37-54.

[22]Jones DK, Horsfield MA, Simmons A, 1999. Optimal strategies for measuring diffusion in anisotropic systems by magnetic resonance imaging. Magn Reson Med, 42(3):515-525.

[23]Jones DK, Knösche TR, Turner R, 2013. White matter integrity, fiber count, and other fallacies: the do's and don'ts of diffusion MRI. NeuroImage, 73:239-254.

[24]Karimi D, Gholipour A, 2022. Diffusion tensor estimation with Transformer neural networks. Artif Intell Med, 130:102330.

[25]Kingma DP, Ba LJ, 2015. Adam: a method for stochastic optimization. Int Conf on Learning Representations, Article 13.

[26]Koppers S, Friedrichs M, Merhof D, 2017. Reconstruction of diffusion anisotropies using 3D deep convolutional neural networks in diffusion imaging. In: Schultz T, Özarslan E, Hotz I (Eds.), Modeling, Analysis, and Visualization of Anisotropy. Springer, Cham, Switzerland, p.393-404.

[27]Lebel C, Walker L, Leemans A, et al., 2008. Microstructural maturation of the human brain from childhood to adulthood. NeuroImage, 40(3):1044-1055.

[28]Le Bihan D, Mangin JF, Poupon C, et al., 2001. Diffusion tensor imaging: concepts and applications. J Magn Reson Imag, 13(4):534-546.

[29]Li HY, Liang ZF, Zhang CY, et al., 2021. SuperDTI: ultrafast DTI and fiber tractography with deep learning. Magn Reson Med, 86(6):3334-3347.

[30]Li ZW, Gong T, Lin ZC, et al., 2019. Fast and robust diffusion kurtosis parametric mapping using a three-dimensional convolutional neural network. IEEE Access, 7:71398-71411.

[31]Lin ZC, Gong T, Wang KW, et al., 2019. Fast learning of fiber orientation distribution function for MR tractography using convolutional neural network. Med Phys, 46(7):3101-3116.

[32]Liu YN, Wang CP, Lu MY, et al., 2024. From simple to complex scenes: learning robust feature representations for accurate human parsing. IEEE Trans Patt Anal Mach Intell, 46(8):5449-5462.

[33]Mori S, Crain BJ, Chacko VP, et al., 1999. Three-dimensional tracking of axonal projections in the brain by magnetic resonance imaging. Ann Neurol, 45(2):265-269.

[34]Nath V, Schilling KG, Parvathaneni P, et al., 2019. Deep learning reveals untapped information for local white-matter fiber reconstruction in diffusion-weighted MRI. Magn Reson Imag, 62:220-227.

[35]Park J, Jung W, Choi EJ, et al., 2022. DIFFnet: diffusion parameter mapping network generalized for input diffusion gradient schemes and b-value. IEEE Trans Med Imag, 41(2):491-499.

[36]Pierpaoli C, Jezzard P, Basser PJ, et al., 1996. Diffusion tensor MR imaging of the human brain. Radiology, 201(3):637-648.

[37]Poulin P, Côté MA, Houde JC, et al., 2017. Learn to track: deep learning for tractography. In: Descoteaux M, Maier-Hein L, Franz A, et al. (Eds.), Medical Image Computing and Computer Assisted Intervention—MICCAI 2017. Springer, Cham, Switzerland, p.540-547.

[38]Roosendaal SD, Geurts JJG, Vrenken H, et al., 2009. Regional DTI differences in multiple sclerosis patients. NeuroImage, 44(4):1397-1403.

[39]Skare S, Hedehus M, Moseley ME, et al., 2000. Condition number as a measure of noise performance of diffusion tensor data acquisition schemes with MRI. J Magn Reson, 147(2):340-352.

[40]Stejskal EO, Tanner JE, 1965. Spin diffusion measurements: spin echoes in the presence of a time-dependent field gradient. J Chem Phys, 42(1):288-292.

[41]Sutskever I, Vinyals O, Le QV, 2014. Sequence to sequence learning with neural networks. Proc 28^th Int Conf on Neural Information Processing Systems, p.3104-3112.

[42]Tian QY, Bilgic B, Fan QY, et al., 2020. DeepDTI: high-fidelity six-direction diffusion tensor imaging using deep learning. NeuroImage, 219:117017.

[43]Tong QQ, He HJ, Gong T, et al., 2019. Reproducibility of multi-shell diffusion tractography on traveling subjects: a multicenter study prospective. Magn Reson Imag, 59:1-9.

[44]Tong QQ, Gong T, He HJ, et al., 2020. A deep learning-based method for improving reliability of multicenter diffusion kurtosis imaging with varied acquisition protocols. Magn Reson Imag, 73:31-44.

[45]Vaswani A, Shazeer N, Parmar N, et al., 2017. Attention is all you need. Proc 31^st Int Conf on Neural Information Processing Systems, p.6000-6010.

[46]Wang CP, Ma B, Xia ZQ, et al., 2022a. Stereoscopic image description with trinion fractional-order continu- ous orthogonal moments. IEEE Trans Circ Syst Video Technol, 32(4):1998-2012.

[47]Wang CP, Hao QX, Xu SJ, et al., 2022b. RD-IWAN: residual dense based imperceptible watermark attack network. IEEE Trans Circ Syst Video Technol, 32(11):7460-7472.

[48]Westin CF, Maier SE, Mamata H, et al., 2002. Processing and visualization for diffusion tensor MRI. Med Image Anal, 6(2):93-108.

[49]Wu YC, Alexander AL, 2007. Hybrid diffusion imaging. NeuroImage, 36(3):617-629.

[50]Wu ZJ, Wang JC, Chen ZQ, et al., 2024. FlexDTI: flexible diffusion gradient encoding scheme-based highly efficient diffusion tensor imaging using deep learning. Phys Med Biol, 69(11):115012.