[The mathematical approaches to differential diagnostics of acute pharyngeal diseases].
Journal:
Vestnik otorinolaringologii
Published Date:
Jan 1, 2017
Abstract
The objective of the present study was to elaborate the program for differential diagnostics of acute pharyngeal diseases based on the 'ENT-Neuro' artificial neuronal network. The study group was formed by means of sampling patients with acute pharyngeal diseases from a set of case histories of the subjects presenting with acute inflammatory diseases. The data thus obtained were employed to develop the expert system to support the decision making process with the use of the 'ENT-Neuro' artificial neuronal network that allows to carry out diagnostics of various inflammatory diseases of the pharynx including the following nosological entities: paratonsillitis, parapharyngitis, acute tonsillitis, and acute pharyngitis, with the minimal probability of erroneous diagnosis (4%). The proposed expert system makes it possible to choose the optimal treatment strategy for the management of various pharyngeal diseases taking into consideration the severity of a concrete pathology and thereby to reduce to a minimum the risk of the related complications.