Using VR and Phased Imaging to Track Alzheimers Disease Progression in New Ways
Earlier in 2012, Swiss researchers successfully managed to generate detailed three-dimensional, VR-navigatable images of the spatial distribution of amyloid plaques spread throughout the brains of mice who were afflicted with Alzheimer's disease. They were able to do this through a combination of VR software, and a new type of imaging.
Amyloid plaques are characteristic of Alzheimer's disease, and are one of the actual sites of damage. They are accumulations of large numbers of extracellular protein aggregates which crop up in random areas of the brain, disrupting local function. They start appearing early in the progression of the disease, and rapidly multiply in number. As the first step in treatment is identifying the precise location of the problem, such a precise, navigatable, interactive imaging method is an extremely useful tool in the toolbox, both for research and for actual diagnosis of patients. In particular with these plaques, the diseases progression could be accurately tracked if it was possible to scan for the current distribution.
Imaging inside a living brain has been achieved through a wide variety of methods. Modst commonly MRI, fMRI and CT scans are used to image inside a living, human brain. However, these imaging methods can only map the terrain. Without something to lock onto, they cannot determine the composition of specific features. fMRI is slightly different from the others, in that it is an example of a type of imaging which can identify features. Namely it locks onto and tracks the haemoglobin in the blood, tracking the use of oxygen by the brain in real-time. It does this by picking up the subtle distortions of the imaging magnetic field, triggered by the difference between oxygen-poor hemoglobin and oxygen-rich hemoglobin.
Any other imaging method that seeks to identify specific structures in the brain must use a similar trick; it must be capable of detecting some aspect about the target area that makes it different from the surrounding areas, in a manner that affects the scanning method itself. You cannot reliably use visual tricks such as analysing terrain contours, the affected areas are buried deep inside a relatively large three dimensional structure with many unique features in different areas.
The researchers made use of an imaging method known as Phase Contrast Imaging. This particular imaging method takes advantage of differences in the refractive index of different substances, to image based on the material difference between the plagues and the surrounding neurons. It is then capable of finding even the smallest plaque deep in the brain, by differences between the types of material. The primary difficulty is then in imaging the brain from every possible angle, and computationally working out whether the refractive differences detected are caused by a single plaque, or multiple plaques, and to attempt to create a 3D model of the living brain with every plaque of the correct shape and in the proper place.
A probable future use for this diagnostic method would be to overlay the plaque map on a map of the brain regions; as we understand more about what processes occur precisely where in the brain, we can compare the plaque locations and determine exactly which mental processes are affected.
The imaging method is a fantastic leap forwards when combined with the computational 3D reconstruction. As one of the researchers involved, Bernd Pinzer, from the Paul Scherrer Institute, explained. Until now, for such an investigation, the brain had to be cut into slices and the slices coloured so that the plaques became visible. This process is the gold standard amongst such investigations. It is, however, very time-consuming, as everything has to be done by hand. At the same time, it provides much less information than our new method. Naturally, we compared the results from our new method with those obtained using this traditional method, and they showed excellent agreement.
The other minor problem with the gold standard method of course, is that opening the skull and slicing up the brain tends to kill the patient. This method doesn't have that problem, and the only brain that is sliced, is a high-detail model entirely held in a virtual reality environment.
As a first concrete result, the researchers determined the distribution of plaques in the brains of a number of mice with different stages of the disease. For each brain, the scientists obtained a three-dimensional image of the overall plaque distribution so that the development of the disease could be followed in detail. With conventional processes, it would hardly have been possible to gather such comprehensive information.
One goal is to use the phase contrast technique to help improve imaging methods which make visible the plaques in the brain of a living patient, and thereby allow a reliable diagnosis of Alzheimers disease to be made, explains Pinzer. These methods are under constant development and it is important to compare their results with those achieved using a known and reliable method. Now it will be possible to directly compare the two sets of 3D images of a mouse brain produced both by a diagnostic method and by our phase contrast technique. One of the diagnostic methods available is Positron Emission Tomography (PET), in which special molecules are attached to the plaques and, after some time, emit gamma radiation, which can be ascertained externally.
A key difference is the new method does not require ingestion of a radioactive isotope which is quite highly radioactive in order for even a small amount to be detected when it is split up and carried by the bloodstream all around the body, until it deposits in the plaques. This radioactive source stays in the body for quite some time afterwards. It is relatively expensive to acquire, takes time to work, and there is a high probability of tissue damage from the internal exposure.
Phase Contrast Imaging doesn't have this problem. No radioactive isotope is used. Simply the imaging tool and the patient. As such there is no waiting period for an isotope to be absorbed, and in theory at least, the imaging method can be carried out on the same day. In fact, it is very similar to an X-ray. The prototype exists at the Swiss Light Source (SLS) at PSI. The SLS generates synchrotron light X-rays that are very intensive and tightly focused. The tight focussiing is key, as it allows the minute deflections caused by different materials in the brain to be detected. A conventional X-ray is just too diffuse to detect these differences.
This tool will allow much more precise studies on how amyloid plaques are distributed, explained Matthias Cacquevel, one of the authors at the École Polytechnique Fédérale de Lausanne (EPFL) , one of the contributing institutions. The relationship between plaques and the symptoms of the disease are still unclear, and information on how these plaques spread throughout the brain is also missing.
So the method offers a way to track the progress of the disease over time, and compare it to the symptoms. Even working with mice, this would be a fantastic step forwards. The virtual reality element of the system is also of great use here. Because the 3D model of the brain is imaged directly into VR as a basic function of the process, it becomes relatively easy to import multiple brain models into the same VR. Models of the same individual brain taken over some period of time. These can then be interacted with such as being layered over one another to track the precise distribution, size and shape changes of plaques in the brain probably using the native capabilities of the diagnostic system itself.
brain amyloid deposition using grating-based differential phase contrast tomography