Track density imaging using diffusion tensor imaging data from 1.5 T MRI scanner


Creative Commons License

Köşker F. B., Gümüş K., TOKMAKÇI M., Acer N., Şenol S., Bi̇Lgen M.

Turkish Journal of Electrical Engineering and Computer Sciences, vol.30, no.6, pp.2044-2053, 2022 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 30 Issue: 6
  • Publication Date: 2022
  • Doi Number: 10.55730/1300-0632.3923
  • Journal Name: Turkish Journal of Electrical Engineering and Computer Sciences
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Academic Search Premier, Applied Science & Technology Source, Compendex, Computer & Applied Sciences, INSPEC, TR DİZİN (ULAKBİM)
  • Page Numbers: pp.2044-2053
  • Keywords: Diffusion tensor imaging, track density imaging, superresolution, tractography, BRAIN, TRACTOGRAPHY
  • Kayseri University Affiliated: No

Abstract

© TÜBİTAK.Superresolution track density imaging (TDI) has recently been developed for achieving high resolution track density maps from low-resolution diffusion images acquired at 3 T. But, the utility of the approach is still unclear when applied to diffusion tensor imaging (DTI) data acquired at lower 1.5 T magnetic field strength and thus its advantages or disadvantages awaits for exploration. We implemented an algorithm to generate track density maps of human white matter using streamline tracking and tested its performance with data acquired from two healthy volunteers at 1.5 Tesla. The effects of number of diffusion directions and seed selections on the quality of the reconstructed TDI maps were investigated under a variety of settings. The results were visually evaluated by an anatomist and a radiologist, and statistically characterized using gray level cooccurance Matrices (GLCM). Producing high-quality maps with improved resolution required increasing the number of seeds per voxel. Statistical implications were consistent with visual inspection. Low signal-to-noise ratio in DTI data intrinsically yielded low SNR in the final TD map. Accurately defining the diffusion and thus fiber orientation within a voxel necessitated increasing the number of diffusion encoding directions. Our data suggests that TDI image with DTI data acquired at 1.5 T is possible using right trade-offs in data acquisition and processing and has the capability of delineating the substructures of the brain.