New patterns recognition of brain mapping based on non-extensive statistical systems

• Grant number: 18/02612-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Effective date (Start): April 01, 2018
• Effective date (End): July 31, 2019
• Field of knowledge: Physical Sciences and Mathematics - Computer Science
• Principal Investigator: Paulo Sergio Silva Rodrigues
• Grantee: Ricardo Morello Santos
• Host Institution: Centro Universitário FEI (UNIFEI). Campus de São Bernardo do Campo. São Bernardo do Campo , SP, Brazil

Abstract

Under the objective of advancing the area of mapping of brain regions, still with many unknown locations, the Human Connectome Project (HCP) is an initiative in this direction, in which it intends to use state-of-the-art scanners for the in vivo mapping of brain regions via magnetic resonance imaging (MRI, fMRI) and computed tomography (CT), on the order of 7 Tesla. With the emergence of this type of strategy, together with free access information, the mapping of human brain structures, mainly in vivo, has improved considerably in recent decades. However, this type of mapping is still far from approaching the levels obtained histologically in the post mortem state. One of the possible solutions is through new techniques of Digital Image Processing and Computational Vision, in order to favor the automatic or semi-automatic mapping of the collected images. Currently, new proposals for segmentation, pattern recognition and analysis of characteristics not only in images but also in videos have appeared in the scientific literature, mainly applied to the medical area. One of the new initiatives that has been causing a great interest of researchers is to attribute to this type of image, or to the morphological or functional structure of the encephalon, a non-extensive character. The theory of non-extensive physical systems is a branch of statistical mechanics that has been considered a generalization of the theory of classical extensive systems, allowing greater flexibility in the description of related phenomena. This new theory includes the reformulation of several well-known physical concepts such as entropy, Gaussian distribution and Fourier analysis, and has been studied with great enthusiasm in several areas of application, especially in the medical field. The present project of Scientific Initiation proposes the use of the theory of non-extensive systems for analysis and processing of images of the HCP project, seeking mainly greater precision in the new mapping that is being proposed.