БАГАТОФАКТОРНЕ ІЄРАРХІЧНЕ НЕЙРОМЕРЕЖЕВЕ РОЗПІЗНАВАННЯ ЗАХВОРЮВАНЬ В УРОЛОГІЇ

The article offers a multifactorial hierarchy structure of neural network modules, developed to diagnose the urologic diseases by processing versatile heterogeneous parameters. These parameters are obtained by carrying out uroflowmetry in patients, using a uroflometer. Based on the results of the analysis, the key parameters for detecting aberrations, obstructions, and diagnosing diseases have been selected. The key feature of the developed multifactorial hierarchy model is the modularized system for detecting the heterogeneous and versatile uroflowmetric parameters. This system is based on neural network modules of different architecture, and training methods. The ability of the neural network model to detect the uroflowmetric parameters in patients has been tested.