Intelligent information technology of diagnostics neuronal processes based on eye-tracking data

Pavlenko V.D., Pavlenko S.V.

Odessa National Polytechnic University, Ukraine, 65044, Odessa, Shevchenko av. 1. Tel.: +3 (0063) 461–74–72,

An intelligent information technology for diagnosing the States of neural processes based on nonparametric identification of the oculomotor system (OMS) in the form of nonlinear dynamic Volterra models is proposed. The technology involves a consistent solution of the following tasks.

1. Identification of OMS. The goal is to construct an information model of OMS in the form of multidimensional transition functions (MTF) – integral transformations of Volterra kernels. Stages of implementation: submission of test signals with different amplitude to the inputs of OMS (horizontal, vertical, diagonal); measurement of OMS responses to test signals using an eye-tracker; calculation of MTF based on the data of the experiment «input-output».

2. Construction of diagnostic model of GDS. The purpose – formation of the feature space. Stages of implementation: compression of MTF; determination of the diagnostic features; selection of optimum system features (reduction of the diagnostic model).

3. Construction of the classifier of the individual's psychophysiological state on the basis of the OMS model. The goal is to build a family of decision rules for optimal classification. Stages of implementation: construction of decision rules based on the results of OMS identification (training); assessment of classification reliability (examination); optimization of diagnostic model.

4. Diagnosis of neural processes. The goal is to assess the state of the individual. Stages of implementation: identification of OMS; evaluation of diagnostic features; classification-classification of the individual under study to a certain class.

Developed software tools that implement all stages of the proposed diagnostic technology.

© 2004 Designed by Lyceum of Informational Technologies №1533