**Using Magnetic Resonance Imaging in the Early Detection of Alzheimer's Disease**

Emily J. Mason, Manus J. Donahue and Brandon A. Ally

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/54445

## **1. Introduction**

Alzheimer's disease (AD) is the most common form of dementia. While many strides have been made in elucidating the underlying causes of AD, studying the disorder *in vivo* has faced several hurdles: First, the structures affected by AD lie deep within the brain where biopsy is not practical. Second, animal models do not develop AD naturally, and genetically engineered models designed to mimic AD do not fully reproduce the human phenotypes [1-3]. Third, while studies using Positron Emission Tomography (PET) have been very use‐ ful for examining plaques and metabolic changes, they involve the injection of radioactive contrast agents. Many of these materials have short half-lives and must be created on-site, making PET very expensive and difficult to be performed at non-specialized centers. Finally, studies which examine cerebrospinal fluid (CSF) require participants to undergo an invasive and sometimes painful lumbar puncture, potentially on multiple occasions [4-7]

In contrast to other techniques, Magnetic Resonance (MR) offers a non-invasive method for analyzing structural and functional brain characteristics without the need for ionizing radia‐ tion. In other words, it can be performed in longitudinal studies without significant health concerns. Multiple scans can be performed quickly in the same testing session to assess tis‐ sue response to tasks or pharmacological administration. The scans are generally 2-5 mi‐ nutes each and many analyses can be done post-hoc. Conveniently, most hospitals and clinics already possess the MR scanners at field strengths of 1.5 and 3.0 Tesla (T).

Many MR techniques have been used to understand the underlying pathology in patient populations already diagnosed with AD. Because MR studies require absolute stillness for several minutes, and some functional scans require the patient to focus on perform‐

© 2013 Mason et al.; licensee InTech. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2013 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

ing a difficult task, performing MR work in advanced AD cases is quite challenging and as such, most studies are limited to mild and very mild cases. While these studies are typically performed at a time when pathology is irreversible, the results of this work point to changes that may be apparent before cognitive decline has become clinically ap‐ parent. For this reason, studies that examine differences between people who will even‐ tually develop AD and people who will not develop AD provide insight into both the cause and the physiology of the disease.

Due to the large amount of water that constitutes tissue (~80-99% depending on tissue type), most MR is specifically focused on the protons on water molecules. Protons have an intrinsic spin that in nature is oriented randomly. In the presence of a magnetic field however, these spins align themselves on average parallel or antiparallel to the axis of the field (Figure 1a). The number of protons aligned parallel to the field is very slightly larger than the number of protons aligned antiparallel, and it is this difference that produces the net magnetization vector in a voxel. When a radiofrequency (RF) pulse is applied at the proper frequency (Lar‐ mor frequency), the longitudinal (z) component of the magnetization vector is tipped away from the axis of the main magnetic field, but continues to spin around the longitudinal axis or "precess" (Figure 1b). When the pulse is removed, the longitudinal component of the magnetization vector will realign itself with the field with a unique time constant that varies

Manipulating the timing of the RF pulses controls the magnetization and creates the desired contrast. The most fundamental timing parameters of relevance are repetition time (TR), echo time (TE), and in some cases inversion time (TI). TR is the time between consecutive acquisitions, and TE is the time from the onset of the excitation pulse that is used for prepar‐ ing the signal for detection to the signal refocusing and in most cases acquisition. In an in‐ version recovery pulse sequence, TI refers to the time between the inversion pulse and the excitation pulse. Importantly, simply by manipulating the timing of the above parameters a range of MR contrasts can be obtained with varying sensitivity to different tissue types. A

by the vector B0 (see Figure 1), and lies along the Z-axis (generally from foot to head).

commonly used, though 1.5T and 3T scanners predominate the clinical setting. The magnet's strength and direction is represented

Using Magnetic Resonance Imaging in the Early Detection of Alzheimer's Disease

http://dx.doi.org/10.5772/54445

227

Due to the large amount of water that constitutes tissue (~80-99% depending on tissue type), most MR is specifically focused on the protons on water molecules. Protons have an intrinsic spin that in nature is oriented randomly. In the presence of a magnetic field however, these spins align themselves on average parallel or antiparallel to the axis of the field (Figure 1a). The number of protons aligned parallel to the field is very slightly larger than the number of protons aligned antiparallel, and it is this difference that produces the net magnetization vector in a voxel. When a radiofrequency (RF) pulse is applied at the proper frequency (Larmor frequency), the longitudinal (z) component of the magnetization vector is tipped away from the axis of the main magnetic field, but continues to spin around the longitudinal axis or "precess" (Figure 1b). When the pulse is removed, the longitudinal component of the magnetization vector will realign itself with the field with a unique time constant that varies with the local environment.

Manipulating the timing of the RF pulses controls the magnetization and creates the desired contrast. The most fundamental timing parameters of relevance are repetition time (TR), echo time (TE), and in some cases inversion time (TI). TR is the time between consecutive acquisitions, and TE is the time from the onset of the excitation pulse that is used for preparing the signal for detection to the signal refocusing and in most cases acquisition. In an inversion recovery pulse sequence, TI refers to the time between the inversion pulse and the excitation pulse. Importantly, simply by manipulating the timing of the above parameters a range of MR contrasts can be obtained with varying sensitivity to different tissue types. A simple pulse sequence indicating RF and

Figure 1. Physics underlying magnetic resonance. Hydrogen atoms align parallel and antiparallel to a strong magnetic field, producing a net magnification vector (a). When a radio frequency pulse is applied at the Larmor frequency, the net magnetization vector is tipped away from the main magnetic field (b). Example of a simple pulse sequence showing timing parameters of the application of radio frequency pulse (RF), the onset

In most cases, the detectible signal (S) that is measured in MRI is a combination of three primary factors: water proton density (C;

The two major contrasts or "weighting" that are used for generating contrast are T1 and T2. T1 and T2 are independent measures and reflect different properties of the tissue of interest, with T1 governing the Mz term and T2 the Mxy term in Eq. 1 above. The time it takes for the magnetization to realign itself longitudinally is measured using T1 weighting (Figure 2a), and is achieved with a short TR and a short TE sequence. T1 is a constant that is unique for each tissue type and is equal to the point when 63% of longitudinal magnetization is recovered (Figure 2b). At the times selected for T1 imaging, there is a high amount of contrast between gray and white matter and therefore T1 weighted imaging is useful for viewing structural changes in the brain (Figure 2c).

ml tissue /100 ml water ), magnetization in the longitudinal plane (MZ) and magnetization in the transverse plane (MXY):

simple pulse sequence indicating RF and gradient timing is illustrated in Figure 1c.

of gradients in the Z direction (Gz), and the timing of signal acquisition (Signal) (c).

**Figure 1.** Physics underlying magnetic resonance. Hydrogen atoms align parallel and antiparallel to a strong magnetic field, producing a net magnification vector (a). When a radio frequency pulse is applied at the Larmor frequency, the net magnetization vector is tipped away from the main magnetic field (b). Example of a simple pulse sequence show‐ ing timing parameters of the application of radio frequency pulse (RF), the onset of gradients in the Z direction (Gz),

gradient timing is illustrated in Figure 1c.

(1) ܯ ή ܯήܥןܵ

and the timing of signal acquisition (Signal) (c).

with the local environment.

It is impossible to predict with certainty who will develop AD, but there are several fac‐ tors that increase the risk. These at-risk populations include individuals in the prodromal stage of AD, termed amnestic mild cognitive impairment (aMCI), and people at a genetic risk for developing AD. A diagnosis of aMCI indicates that there is more memory de‐ cline than would be expected based on the person's age and education level, however memory impairment is not interfering with daily activities. It is estimated that 10-20% of people 65 and older have aMCI, and out of those 10-15% will progress to develop AD in 3-4 years. [8,9] Because approximately 30% of people diagnosed with aMCI will remain stable or improve over time, it is important to find biomarkers that will identify those most likely to progress to AD.

This chapter will focus on the use of MR in the early detection of AD. Major advances have been made in structural imaging of both gray and white matter using proton density, T1 and T2- weighted imaging, and Diffusion Tensor Imaging (DTI). Functional imaging in AD will also be reviewed, and Blood Oxygenation Level-Dependent (BOLD) functional Magnet‐ ic Resonance Imaging (fMRI) will be broken down into its primary contributors: Cerebral Blood Flow (CBF), Cerebral Blood Volume (CBV), and the Cerebral Metabolic Rate of Oxy‐ gen (CMRO2). Finally, hemodynamic fMRI contrast can be complemented using measures of neurochemistry, including measuring the balance between excitatory (glutamatergic) and inhibitory (γ-aminobutyric acid; GABAergic) neurotransmission. This can be achieved with new single-voxel chemical imaging techniques such as Magnetic Resonance Spectroscopy (MRS), or more recently using multi-voxel MRS imaging (MRSi)
