Scalable joint segmentation and registration framework for infant brain images

Abstract

The first year of life is the most dynamic and perhaps the most critical phase of postnatal brain development. The ability to accurately measure structure changes is critical in early brain development study, which highly relies on the performances of image segmentation and registration techniques. However, either infant image segmentation or registration, if deployed independently, encounters much more challenges than segmentation/registration of adult brains due to dynamic appearance change with rapid brain development. In fact, image segmentation and registration of infant images can assists each other to overcome the above challenges by using the growth trajectories (i.e., temporal correspondences) learned from a large set of training subjects with complete longitudinal data. Specifically, a one-year-old image with ground-truth tissue segmentation can be first set as the reference domain. Then, to register the infant image of a new subject at earlier age, we can estimate its tissue probability maps, i.e., with sparse patch-based multi-atlas label fusion technique, where only the training images at the respective age are considered as atlases since they have similar image appearance. Next, these probability maps can be fused as a good initialization to guide the level set segmentation. Thus, image registration between the new infant image and the reference image is free of difficulty of appearance changes, by establishing correspondences upon the reasonably segmented images. Importantly, the segmentation of new infant image can be further enhanced by propagating the much more reliable label fusion heuristics at the reference domain to the corresponding location of the new infant image via the learned growth trajectories, which brings image segmentation and registration to assist each other. It is worth noting that our joint segmentation and registration framework is also flexible to handle the registration of any two infant images even with significant age gap in the first year of life, by linking their joint segmentation and registration through the reference domain. Thus, our proposed joint segmentation and registration method is scalable to various registration tasks in early brain development studies. Promising segmentation and registration results have been achieved for infant brain MR images aged from 2-week-old to 1-year-old, indicating the applicability of our method in early brain development study.

U-ReSNet: Ultimate Coupling of Registration and Segmentation with Deep Nets

“…In [11] the authors present a graph-based method applied on MR brain images where the registration constraints are relaxed in the presence of brain tumors, while their formulation provides also segmentation masks of the tumor area. Using ultrasound images, a joint variational framework for the joint cosegmentation and registration has been proposed in [13], while in [5] the authors address both tasks in a joint framework, evaluating them on infant brain images. Even though there is a wide range of methods addressing image segmentation and registration jointly, showing that one solution impacts the other, according to our knowledge, this is the first time that an efficient formulation based on deep learning is presented and evaluated on brain MR images.…”

Application of Partial Differential Equation in Brain Image Segmentation and Registration

“…Image segmentation is a key basic operation in image processing, which provides a basis for image recognition and understanding. The problem of image segmentation exists with the generation of image [16]. Up to now, there have been many image segmentation methods, but each method has its specific scope of application.…”

“…Many researchers have proposed a large number of processing methods [15]. However, because of the complexity of image content and the different needs of users, each method is usually only applicable to specific images or needs, so these two problems are still the research hotspots in image engineering [16]. In addition, based on the interrelationship between registration and segmentation, researchers have proposed a coincidence model to realize registration and segmentation at the same time, and the exploration of coincidence method has become a hot issue in recent image processing [17], [18].…”

Brain Image Segmentation Based on Multi-Weight Probability Map

“…However, considering the difference in human anatomy, single-spectral segmentation is difficult to adapt to the differences in brain structure of different individuals, and it is easy to produce false segmentation, especially at the boundary. Segmentation methods based on multi-spectral registration [9] use multiple sets of medical maps to reduce the uncertainty of map selection. The multi-map registration method based on multimap registration uses multiple sets of maps, so a fusion algorithm is needed to fuse multiple sets of segmentation results together.…”