In a first step, you will need to define the brain structure in which you are interested. Basically, you need to label the pixels (or voxels, in the case of 3D imaging) of your dataset, depending on whether they belong the respective brain structure or not, This process is called segmentation; a wide range of methods exist for this purpose.
The second step is simpler: You need to count how many pixels/voxels belong to your brain structure, and you need to know the volume of each pixel. Calculating the volume of your brain structure is then a simple matter.
A wide range of methods and (semi-)automatic algorithms for the purpose of segmentation has been developed. The choice of algorithm for a particular application depends on numerous factors, such as the available image contrasts (T1, T2, FLAIR, ...), the type of acquisition sequence, the resolution and obviously the structure of interest.
Of course, a range of refinements exist, and a wide range of potential difficulties: How do you deal with partial volume effects - i.e. the case that a single pixel contains several structures? This is particularly relevant when you want to quantify small volumes with a a coarse resolution.
You could try our new online platform for automatic MRI brain analysis.
Our web based pipeline is called "volBrain" and can be accessed at this address:
http://volbrain.upv.es
volBrain is able to produce volumetric information of different brain tissues/structures in a fully automatic way in around 15 minutes. The way it works is very simple; you have just to register as a new user in the system and to upload a single anonymized compressed T1w brain MRI in nifti format. You will receive an email with a pdf/csv summarizing the results and you will be able also to download the resulting label files and images in your user area using your account.
Our aim is to provide a simple and free tool that can be used no matter where you are without any installation or computer requirements. Since we want to make the tool available to the higher possible number of users, a limit of 5 cases per day/user has been set.
We encourage you to register in the system and to use it. We will be very grateful if you can provide us your comments/suggestions in order to improve the system.
Good luck in your research!!
Jose V. Manjon and Pierrick Coupe
4 years after Pierrick Coupé added the answer, it can save my life and days and years. I wonder whether 5 limit per user is still ongoing. I do not have enough time.
- Badges
- Science topic