Abstract
To understand more about the human brain and how it works, it is vital to understand how the neural circuits connect different regions of the brain. The human brain is filled predominantly with water and the majority of the water molecules undergo diffusion which can be captured with the help of diffusion MRI. Diffusion weighted images enable us to reconstruct the neural circuits in a non-invasive manner, and this procedure is referred to as tractography. Tractography aids neurosurgeons to understand the neural connectivity of the patient. This chapter attempts to explain the procedure of tractography and different types of algorithms.
TopBackground
It all started, when Einstein in 1905 did an investigation on the random motion of water molecules. The human brain is filled predominantly with water and also has other fluids such as cerebrospinal fluid. Random motion of water molecules is observed which is referred to as Brownian motion. Fast forward big MRI machines with the help of magnetic fields created dMRI images. In the case of dMRI images, the MRI signal containing diffusion information is sensitized to the direction in which the magnetic field is being applied.
The arrangement of the white matter fibers in our brain gives rise to the diffusion signal and the acquired image is referred to as diffusion-weighted image (DWI). Scientists developed gradient schemes when entered on the MRI machine captures the diffusion behavior of water molecules. There are schemes such as one-shell acquisition, where only one magnetic field is applied, whereas in multi-shell acquisitions more than one magnetic field is used to acquire the dMRI image. For instance, if the dMRI image is acquired using only one magnetic field b = 1000 s/mm2 then it is referred to as one shell dataset and if the dMRI image is acquired using more than one magnetic field such as b = 1000 s/mm2 and b = 3000 s/mm2 then the dataset is referred to as multi-shell acquisition. In the MRI sequence developed for capturing the dMRI image, a large number of gradient directions are defined. More the number of gradient directions more accurately the tractography algorithms can reconstruct the white matter fibers.
In clinical settings, one of the majorly used methods for the reconstruction of white matter fibers is Diffusion Tensor Imaging. In DTI, the tracking of white matter fibers is done by calculating the eigenvectors in each voxel of the brain. Voxel is analogous to what pixel is in a 2-Dimensional image. DTI method has its limitations, which we will see in detail in this chapter and understand why the other methods perform better than DTI.
Key Terms in this Chapter
DTI: DTI refers to diffusion tensor imaging which is a type of deterministic method predominantly used in clinical applications.
Voxel: A 3D unit that embeds the signal in brain scans.
fODF: Defined for all fibers in the unit. Indicates the probability of a fiber in a given voxel.
White Matter Fibers: Areas of the central nervous system that are made up of myelinated axons. Also referred to as tracts.
DWI: DWI refers to diffusion weighted image which is captured with the help of dMRI. Captures the diffusion information of the water molecules inside the brain.
Diffusion: Movement of water molecules from an area of higher concentration to lower concentration.
dMRI: dMRI stands for diffusion Magnetic Resonance Imaging which is a popular method used by neuroscientists to uncover unique information about the neural connections within the brain. Tractography - A 3D modelling technique used to represent the white matter fibers inside the brain.