Patients with Alzheimer's disease (AD) often forget fundamental aspects of their everyday life. This begins with them forgetting more recently made memories or recently met people. Slowly, they then begin to forget things that happened longer ago in their lives. They may forget family members, like children or husbands and sometimes siblings. This disease can be heartbreaking for both the patients who have it and their family members.
Different imaging methods, such as computed tomography (CT), magnetic resonance imaging (MRI) or positron emission tomography and MR combined (PET/MR), can provide insight into the brain to help physicians determine whether a patient has Alzheimer's, as well as how quickly their brain is deteriorating. In order to do this, it is desirable to not only visualize the scans but also quantify a range of image features with may distinguish AD from other forms of dementia.
Quantifying Alzheimer's disease
Studies have begun to show that quantification of measures such as the rate of atrophy can help to distinguish Alzheimer's disease from that of healthy brain aging. For example, a recent study has shown that the median rate of atrophy over one year was greater in the AD group than the control group.1 This could help to lead to better differentiation when considering the disease and could provide more information for the referring physician.
Another form of data leverage used in connection with stratification of AD comes in the form of deep learning, which is currently an area of extensive research.2 This particular approach may compare an individual to an extensive database of expertly labelled images for bench-marking. Deep learning could potentially provide quantified information to the referring physician that can help in making the diagnosis and could potentially aid prognoses or help predict treatment response.
Longitudinal studies of Alzheimer's disease
In order to ensure that the longitudinal study of Alzheimer's disease is as effective as possible, it is important that the scans are repeatable. Some new technology can help to provide more consistent imaging across multiple scanners and technologists.3 This semi-automated technology uses anatomical landmarks to help the technologist select the right slice and angle, which could lead to increased consistency between subjects and sessions. This standardization could be particularly helpful when performing longitudinal assessments to track disease progression.
While many longitudinal imaging assessments have mainly been limited to research studies, there is an ever-increasing need for such measures in clinical trials. MRI can detect localized changes in brain volume in specific regions, which could indicate whether a drug is having a disease modifying effect. The treatment may be neuroprotective, which could lead to the prevention or slowing down of progressive atrophy or potentially facilitating neuroregenerative processes. If doctors are able to objectively measure the central effect of the potential treatment, then this can inform the patient's clinical management.
Distinguishing Alzheimer's disease
Some image acquisition techniques and analysis tools may also help to differentiate between Alzheimer's and other forms of dementia. One study found that frontotemporal dementia and Alzheimer's could be distinguished by the spatial distribution of hypoperfusion on arterial spin labeling MRI.4 Combined with structural imaging and the appropriate analytic, these differences can be highlighted and may aid differential diagnosis, which can be especially challenging because the symptoms often overlap. As a result, the patient may receive the most appropriate, personalized assessment, which will guide more precise treatment
Another method that can be used to determine whether a patient has Alzheimer's or a different form of demetia comes from the use of PET.5 It is now possible to co-register and overlay amyloid PET scans onto an MR template to aid visualization of the molecular pathology and cross reference to the underlying brain anatomy. Potential uses of this technique include earlier detection in the at-risk prodromal phase of the disease, which may be more pertinent for prophylactic, neuroprotection strategies.
The effects of Alzheimer's disease are often difficult to observe and measure early on. However, neuroimaging can provide valuable insight into a patient's current neurobiological state and the likelihood of disease progression. Differentiation between AD and other forms of dementia is still being researched, but understanding the locations of the atrophy and topology of pathogens such as amyloid or tau could improve our chances of disease differentiation. Longitudinal studies could also help us to understand the treatment effectiveness and disease progression of Alzheimer's disease. Both differentiation and longitudinal studies rely on the quantification of the atrophy in the brain, which could be helped by consistent scan selection and artificial intelligence approaches to the analysis. Patient stratification can also be complimented by the detection of amyloid proteins, for example. Hopefully, combining this imaging data with clinical history, cognitive tests and biofluid-based biomarkers will improve our understanding of Alzheimer's and support the development of new treatments.
- Visualisation and quantification of rates of atrophy in Alzheimer's disease. The Lancet. https://www.sciencedirect.com/science/article/pii/S0140673696052282. Last accessed September 23, 2019.
- Prediction and classification of Alzheimer disease based on quantification of MRI deformation. PLOS.org. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0173372. Last accessed September 23, 2019.
- Ultra-flexible AIR Technology Suite making a difference in the technologist's workflow. SIGNA Pulse of MR. http://www.gesignapulse.com/signapulse/spring_2019/MobilePagedArticle.action?articleId=1488817&app=false. Last accessed September 23, 2019.
- Hypoperfusion in frontotemporal dementia and Alzheimer disease by arterial spin labeling MRI. Neurology.org. https://n.neurology.org/content/67/7/1215.short. Last accessed September 23, 2019.
- Alzheimer's Disease. SignaMasters. Last accessed September 24, 2019.