CHAPTER 3Overview of Image Recording TechniquesThe field of image recording is an immense and ever-expanding field. As of early 1993 there were over 120 articles written on the registration issue, as cited in the survey article written by van den Elsen et al.[199]. Since then the number of published articles has grown exponentially. This chapter will discuss the elements of recording techniques according to a classification originally proposed by van den Elsen et al.[199] and subsequently extended by Maintz et al.[120]. The set of criteria described is explained in Figure 3.1. This classification includes dimensionality of the algorithm, nature of the registration algorithm, nature and domain of transformation, user interaction, optimization procedure, modalities involved and type of subjects used in the algorithm.3.1 Dimensionality - 2D, 3D, 4 DOne of the most obvious classifications that have emerged from the set of image recording techniques concerns Dimensions, or how many dimensions are used in the recording process. The range of this size can vary from a simple 2D recording to a complex time series recording of 3D data, i.e. a 4D process [49]. Based on the dimension recording algorithm can be divided into two parts: those that deal with time series registration and those that do not deal with, that is, they deal only with the spatial dimension. 3.1.1 Registration involving spatial dimension The algorithm dealing only with this field of spatial dimensions can be further classified into 2D and 3D. The registration algorithm can be applied very easily to both 2D and 3D datasets. The only difference in 3D, however, is that the size of the dataset is significantly increased and the number of transformations... middle of paper... tomographic images are matched to an anatomical atlas or some other model. This procedure can facilitate automatic segmentation [42].2.7 SubjectThis classification refers to the subjects involved in the registration process. This can be classified as intersubject, intrasubject, or subject to pattern registration. Intra-subject is mainly used. In this data comes from the same scenario, as from the same patient. This type of recording can be used in almost any diagnostic. Intersubject is quite complicated since the transformation must overcome the inherent anatomical differences that exist between two different scenarios. That's why the intersubject algorithm is mainly based on curved transformation. The subject to the model is essentially the same as the modality to the model. Therefore all techniques related to the modeling modality are also applicable to subjects subject to model registration.