The chapter presents an overview of the server subsystem of the PsyScan platform, designed for self-diagnosis of depressive states and to assist doctors in decision-making. The focus is on the description of the architecture and functional capabilities of the server component, which ensure the collection, processing, and storage of user data. The subsystem leverages modern machine learning approaches and natural language processing techniques to analyze information from surveys, patient records, and other sources. Special attention is given to data protection measures and user privacy, which are critical in medical applications. The chapter provides testing results that demonstrate the subsystem's efficiency and reliability, as well as discusses prospects for its further development within the platform for depression diagnosis.