Redes cerebrais na depressão: um estudo das propriedades topológicas

Abstract

Many natural systems are composed of multiple interacting elements whose dynamics cannot be explained by the simple sum of their parts. These so-called complex systems require analytical tools capable of capturing both structural and functional relationships among their components. In this context, complex network theory has emerged as powerful framework to model such systems, bridging mathematical, physical, biological, and computational concepts. Among known complex systems, the human brain stands out as one of the most emblematic. Its functional and structural organization can be represented as networks in which different brain regions interact through anatomical (structural) connections and temporal (functional) correlations. Scientific evidence indicates that healthy brains exhibit small-world topology, marked by high local clustering and efficient global integration, which supports communication across regions at low connection cost. This work begins with a systematic literature review on the topological organization of the brain in individuals with depression. The review reveals significant divergence in the reported results regarding connectivity metrics and small-world properties. While some studies report a weakening of network organization, others suggest opposite effects, such as topological strengthening or randomization. To address these inconsistencies, we propose an exploratory approach based on computational simulations of small-world networks using the Watts-Strogatz model. The aim is to investigate how variations in topological parameters affect classical integration and segregation metrics. These simulations help illustrate the potential impacts of different network configurations on brain organization, offering theoretical support for critically interpreting empirical findings in depression neuroscience.