Scientists from the Baltic Federal University. Immanuel Kant (Kaliningrad) and Plovdiv Medical University (Bulgaria) proposed an interpretative approach that allows to effectively detect depressive disorder with an accuracy of 83%. The results of the research were published in a scientific journal. Chaos: Journal of Interdisciplinary Nonlinear Science.
As part of the study, the scientists used images obtained using functional magnetic resonance imaging, a method based on measuring the density of blood flow. When any area of the brain was stimulated, blood flow to it increased. Then, based on the resulting images, the researchers created graphs – complex networks where nodes (dots) mimic brain regions, and connections between nodes – graph edges – correspond to interactions between regions. Based on the resulting graphs, the scientists compared the brain function of 35 patients with depressive disorder and 50 healthy people, and then tried to separate the feature space of both groups using machine learning methods.
The proposed approach made it possible to distinguish between sick and healthy people with an accuracy of 82.6%. At the same time, the groups differed better if the authors took into account a specific set of network metrics in the analysis: the strength of the node corresponding to the activity of the brain region, the number of edges, that is, the number of interactions between regions and the clustering coefficient, which indicates the tendency of nodes to cluster together. If the researchers applied the features separately or added other unimportant network features, the algorithm did not work correctly.
Thus, the method made it possible to reveal features of the functional network of the brain that differ in sick and healthy subjects. At the same time, the algorithm recognized differences in the overall properties of the entire neuron network, not differences in individual local connections between brain regions (as is done in most methods). The new technique will allow monitoring of global changes in the structure of the brain in people with depressive disorder and will provide clinical application with a reliable way to diagnose this disease.
“In the future, based on the data obtained, we plan to highlight the characteristic features of the functional networks of the brain of healthy people and patients with major depressive disorder at different levels of the brain. Project participant Andrey Andreev, Senior Researcher of the Baltic Federal University, Candidate of Physical and Mathematical Sciences, this is to determine the disease based on magnetic resonance imaging will allow us to select the most important biomarkers to be analyzed for Immanuel Kant.
Major depressive disorder is a mental illness that affects approximately 280 million people worldwide. Patients lose interest in activities, face insomnia, lethargy, weakness, guilt and self-humiliation.
ancient biologists to create A gene responsible for impaired ability to enjoy in depression.