Biological sciences /11.Bioengineering and bioinformatics

Dr.Sc.(Technology) Semenets V.V. , Cand.Sc.(Technology) Natalukha Yu.V, Cond.Sc. (Medicine) Taranukha О.А., Cand.Sc. (Technology) Tokarev V.V.

_{Kharkov National University of Radio Electronics (KNURE)},

Kharkov, Ukraine

_{Kharkov National Medical University} (KNMU),

Kharkov, Ukraine

_{Kharkov National University of Radio Electronics (KNURE)},

Kharkov, Ukraine, tvv.v@mail.ru

Computer system for retina non-invasive diagnostics of human visual analyzer

Particularity of the retina diseases of the human visual analyzer consists in that they often cause persistent loss of ability to work. Substantial growth in disability due to loss of vision is observed among people of the working age, the reasons for this are as follows: ischemic optic neuropathies, glaucoma and visual analyzer traumas. The result is the need to develop the computer system (CS) of noninvasive methods for the predictive diagnostic of the human visual analyzer retina aimed at diseases detection in the early stages. The developed computer system (CS) for the retina noninvasive diagnostics of the human visual analyzer can read and process automatically the video information, store the received video data, diagnose retinal diseases of the human visual analyzer and exclude the influence of human factors on the survey results. Processing of video information, needed to diagnose retinal diseases of the human visual analyzer, is carried out in the proposed computer system on the basis of fast precision algorithms and hardware and software for measurement of the retina image informative parameters. Initially the image pre-processing is carried out, namely, digital filtering, noise reduction, elimination of glare from an active illumination of an eye and other interferences. In many occasions the distortions can approximately be considered to be the consequence of the original signal linear transform. This occurs, for example, as a result of the atmosphere turbulence, movement or aberrations of the optical system. Another feature of the observed image is the presence of additive interferences (noise) in it. Noises occur in the paths of generation, transmission and reception of signals. Recently, linear methods of images restoration (IR), used in the spatial and frequency domains, have found an extensive application. Thus, both deterministic and statistical approaches are used for describing two-dimensional signals (fields). Let and be the intensity distribution in the image plane and the original plane, respectively, - the point scattering function (PSF) on the assumption that the display is the linear one. The general equation to describe has the following form [3]:

, (1)

where - is the additive random noise.

For the spatial-invariant function of the scattering point = and the integral in the right-hand side of equation (1) leads to the two-dimensional convolution, i.e. [3]:

, (2)

Taking the form of the distortions described by expressions (1) or (2), the restoration problem can be formulated as follows: with the known_, find a good estimate of ( denoted as having a priori information concerning the values of _. The applied methods for restoration of images based on the finite de-convolution windows use different number of known a priori information and various quality criteria of assessment of. The developed computer system for retina non-invasive diagnostics of the visual analyzer makes it possible to carry out image capture, analysis and measurement of the retina in real time (fig. 1). The specialized software can realize the automatic mode of operation of computer theme, adaptation for the conditions of control and observation, processing of video information and measurement of integral informative parameters of images.

Figure 1 – Image of retina visual analyzer physiological norm

Methods of redundant information reduction and increase in the computer systems speed of operation in the information readout processes themselves can be used in multi-element transducers with a random signals’ sampling. Fig. 2 demonstrates the block diagram of the computer system for retina non-invasive diagnostics of human visual analyzer.

Figure 2 – Block diagram of computer system for retina non-invasive diagnostics of human visual analyzer

The developed CS makes it possible to increase the precision of measurement and control of the retina visual analyzer informative parameters and thus to increase the reliability of diagnostics of congenital or acquired pathologies, in its turn this will allow to increase the efficiency of the disease treatment or prevent them in the initial stage of development.

Literature:

1. Computer method of investigation into violation of color sense. Shamshinova А.М., Petrov А.S., Dvoryanchikova А.P., Arefyeva Yu.А., Eskina E.N., Kiselyeva Т.N., Zol’nikova I.V. // Bulletin of Ophtalmology - 2000. – V. 116. - №5. - P. 49-51. (In Rus.)

2. Functional methods of investigation in ophthalmology / Shamshinova А.М., Volkov V.V. - М.: Medicine, 1999. - 416 p. (In Rus.)

3. Tokarev V.V. Image restoration based on finite de-convolution windows [Text] / V.А. Afanas’iev, М.А. Il’in, Yu.V. Natalukha, V.V. Tokarev // Bionics of Intelligence.- 2012. - №2 (79). – P.80-83. (In Rus.)

4. Tokarev V.V. On one method of non-invasive diagnostics of macular area of retina human visual analyzer [Text] / V.V. Semenets, Yu.V. Natalukha, О.А. Taranukha, V.V. Tokarev // APLIKOVANÉ VĚDECKÉ NOVINKY - 2014: мateriály X - mezinárodní vědecko - praktická konference 27 červenců - 05 srpna 2014 roku.- Praha, 2014. - P.67 - 69.

5. Tokarev V.V. Visual analyzer model in computer system for diagnostics. [Text] / V.V. Semenets, Yu.V. Natalukha, О.А. Taranukha, V.V. Tokarev // Information systems and technologies: Proc. of III^rd International Scientific Conf., September, 15-21, 2014 г. – Kharkov, 2014. – P.81-82. (In Rus.)