High‐resolution and high‐fidelity diffusion tensor imaging of cervical spinal cord using 3D reduced‐FOV multiplexed sensitivity encoding (3D‐rFOV‐MUSE)

• Published in: Magnetic resonance in medicine Vol. 94; no. 1; pp. 166 - 182
• Main Authors: Yuan, Chenglang, Chen, Shihui, Liang, Liyuan, Xu, Xiaorui, Xiong, Hailin, Li, Yi, Liu, Tianbaige, Chen, Nan‐Kuei, Chang, Hing‐Chiu
• Format: Journal Article
• Language: English
• Published: United States Wiley Subscription Services, Inc 01.07.2025; John Wiley and Sons Inc
• Subjects: 3D diffusion MRI; Adult; Algorithms; Biomarkers; Cerebrospinal fluid; Cervical Cord - diagnostic imaging; cervical spinal cord; Coding; Data acquisition; diffusion tensor imaging; Diffusion Tensor Imaging - methods; Feasibility studies; Female; Ghosts; Healthy Volunteers; high‐resolution DTI; Humans; Image Processing, Computer-Assisted - methods; Image reconstruction; Imaging Methodology; Imaging, Three-Dimensional - methods; Magnetic resonance imaging; Male; multiplexed sensitivity encoding; Multiplexing; MUSE; reduced‐FOV; Sensitivity; Signal-To-Noise Ratio; Spinal cord; Spinal Cord - diagnostic imaging; Tensors; high‐resolution DTI; cervical spinal cord; multiplexed sensitivity encoding; reduced‐FOV; 3D diffusion MRI; diffusion tensor imaging; MUSE
• ISSN: 0740-3194; 1522-2594; 1522-2594
• DOI: 10.1002/mrm.30455

Purpose To develop a 3D isotropic high‐resolution and high‐fidelity cervical spinal cord DTI technique for addressing the current challenges existing in 2D cervical spinal cord DTI. Methods A 3D multi‐shot DWI acquisition and reconstruction technique was developed by combining 3D multiplexed sensitivity encoding (3D‐MUSE) with two reduced FOV techniques, termed 3D‐rFOV‐MUSE, to acquire 3D cervical spinal cord DTI data using a sagittal thin slab. A self‐referenced 2D ghost correction method and a 2D navigator‐based inter‐shot phase correction were integrated into the reconstruction framework to simultaneously eliminate Nyquist ghost and aliasing artifacts. Cardiac triggering was used during data acquisition to minimize the influence of cerebrospinal fluid pulsation. In vivo experiments were conducted on five healthy subjects at a 1.5 T MRI scanner for evaluating the feasibility of 3D cervical spinal cord DTI using 3D‐rFOV‐MUSE in terms of geometric fidelity, reconstruction performance, and SNR efficiency. Results A 3D‐rFOV‐MUSE can achieve high‐resolution cervical spinal cord DTI at 1.0 mm isotropic resolution. The integration of reduced FOV and multi‐shot acquisitions can improve the geometric fidelity of 3D cervical spinal cord DTI. Compared with routine 2D single‐shot diffusion‐weighed EPI (2D‐ss‐EPI), the proposed technique can mitigate through‐plane partial volume effect and enable multi‐planar data reformation for cervical spinal cord DTI, with effective reductions of distortions and improved signal‐to‐noise ratio. Conclusion We demonstrated the feasibility of high‐resolution and high‐fidelity 3D cervical spinal cord DTI at 1.0 mm isotropic resolution using 3D‐rFOV‐MUSE technique, which may potentially improve the quantitative assessment of cervical spinal cord DTI biomarkers.