[SOUND] Okay, so we'll now start theme four. Advanced Brain Image Processing. We'll cover some of the same topics we covered before. But we will go in a little bit more detailed, we will also cover a few new topics. In particular we will talk about again within-subject registration, ROI localization, and segmentation. However, we will go in a lot more detail and provide more hands on examples in terms of how to actually do this in practice. We'll revisit some of the things about co-registration from this lecture. We'll go back over the definitions, the basic components, and the pipeline tools. You probably remember this slide, where we talked In some detail about the various types of registration. We talked about complexity, co-registration, registration to a template or one subject to another registration. In particular, what we'll cover in these lectures will be rigid six degrees of freedom registration. And we'll look at cross sectional between modalities registration known longitudinal within modality registration and longitudinal between modalities registration. So let's revisit some of the concepts that we've seen before. Remember that we were talking about linear registration, rigid registration and we were saying that there are six degrees of freedom and you'll have the transformation on the slides, T for transformation index rigid because it's a rigid transformation and v and that is equal to rotation matrix R which you have on the slides. Times the location v plus a translation vector t. There are three degrees of freedom in the rotation metrics which are the three angles of rotation and there is a translation there're also three degrees of freedom in a translation vector t for a total 6 degrees of freedom. To get a better idea about what the rotation matrix tries to capture, is actually the pitch, the roll, and the yaw. And here you have a nice image that explains a little bit better what the pitch, the roll, and the yaw is. The pitch essentially refers to nodding yes. Yes, roll means rolling no. And yaw means I don't know. You maybe careful if you may want to be careful because in some countries, yes and no are actually interchangeable. Yes and no actually mean the exact opposite. To get an overall view of what we are trying to accomplish, on this slide, I actually present the complexity of the problem we are trying to solve. On the left side you have three images, which are baseline images, each one of them referring to a different model. In this case where we have the T1 the T2 and the flare. And on the right side of the plot you have the follow-up T1, the follow-up FLAIR, and the follow-up T2. So, these are two visits, each one of them with three different brain images, and there are different types of registration that could be done. This is a good, I think, a good visual representation of the problem, and there are different specific things you can do, you can register the baseline FLAIR and T2 to the baseline T1, which will be one of the things we will be doing in the next lecture. You can also register longitudinally. You can register the follow-up T1 to the baseline T1. Or you can register at the second visit the follow-up FLAIR or the follow-up T2 to the follow-up T1. We'll go in some details with respect to each one of these types of registrations, and we will also see exactly how to do this using FSLR and ANTsR.