Analysis of Medical Images by Computational Methods: Algorithms and Applications
João Manuel R. S. Tavares · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · 2014
The analysis of structures in images by computational methods is a very challenging issue as it frequently involves automatic tasks for segmentation, i.e., the detection of the structures represented, extraction of representative features from the structures, matching between images, rigid and non-rigid alignment of images, temporal tracking and motion analysis of features in image sequences, deformation estimation between two structures, as well as the 3D shape reconstruction of the structures from images. For example, to analyze the behavior of organs from sequences of medical images, first the input images should be segmented, then suitable features of the organs under analysis should be extracted and tracked along the image sequences and finally the motions involved should be tracked and analyzed. Despite the inherent difficulties, computational methods of image analysis provide a wide range of important applications for our society. Applications regarding 2D, 3D or even 4D image data can be easily found in engineering, materials sciences and virtual reality. However, nowadays the most predominant applications of these methods are found in medicine, biomechanics and bioengineering. In this talk, the computational methods of image analysis that we have developed in order to analyze structures from images will be introduced; particularly, those which have been used for image segmentation, matching, alignment, tracking, as well as for 3D shape reconstruction from medical images. Furthermore, their use in applications from medicine and biomechanics will be presented and discussed.