Morphological Characteristics Analysis of Working Memory Tracts Using BOLD-fMRI and HARDI Based Tractography in Healthy Human Brains

• Published in: Bioengineering and Biomedical Signal and Image Processing pp. 267 - 280
• Main Authors: Troudi, Abir, Alaya, Ines Ben, Labidi, Salam
• Format: Book Chapter
• Language: English
• Published: Cham Springer International Publishing 2021
• Series: Lecture Notes in Computer Science
• Subjects: fMRI; HARDI; Tractography; Working memory bundle
• ISBN: 3030881628; 9783030881627
• ISSN: 0302-9743; 1611-3349
• DOI: 10.1007/978-3-030-88163-4_24

The neuroimaging field has newly become a drift with multimodal neuroimaging findings to study brain connectivity through the combination of High Angular Resolution Diffusion Imaging (HARDI) based tractography, and functional activation maps using functional Magnetic Resonance Imaging (fMRI) task-related. The reconstruction of Working Memory (WM) fibers bundle is an important goal for many clinical applications such as brain tumors, gliomas, traumatic brain injury, schizophrenia, and Alzheimer Disease. In the present paper, we propose to study the anatomical variability of the WM bundle by focusing on the quantification of the track and testing for correlations with sex and local reconstruction methods. The dataset used for the experiment is based on in vivo data from the Human Connectome Project (HCP). We used fifty normal volunteers (25 males, 25 females; mean age = 25 ± 2.81 years). We computed the shape similarity, the volume, the mean length, and the number of streamlines, to evaluate the variability of macro structural measures between subjects. Firstly, this study demonstrates that there is a significant difference between the reconstruction of WM fiber bundle using the Diffusion Tensor Imaging (DTI) model and the High Angular Resolution Diffusion Imaging (HARDI) model, with a mean correlation coefficient r of 0.43 and a p-value of 0.09 for all measured parameters. Also, the WM bundle has a complex architecture with crossing fibers. Therefore, we can conclude that the HARDI model is the most relevant model for the reconstruction of the WM fiber bundle. Secondly, there were a slight sex differences in WM fiber bundles morphology between healthy subjects, with a 0.95 shape similarity average. Finally, we find a slight difference between the measured macrostructural parameters inter-subjects (Normalized Volume, Mean Length of streamlines (MLS), and Number of streamlines). We conclude that the statistical measures show that the MLS is the well-correlated parameter.