Ilyukhin Evgeny

Theme of dissertation: " Development of specialized computer system of processing of tomograms of neoplasms of a liver "
Supervisor: Ph. P. Jaroshenko Nikolay Aleksandrovich
My e-mail: iluhin@sktel.com.ua

ABSTRACT

Each newest discovery in the physics or technics inevitably finds an embodiment in medicine; as a vivid example the discovery of Roentgen and its introduction in a medical practice can serve that. The computer tomography (CТ) can be considered as a new coil in development of radiology, in turn, principles of mathematical processing at construction of the image at CТ have laid down in a basis of last modern development.

CТ Data can be used for carrying out of a diagnostic puncture, and, that is especially important, it can be applied with success not only to revealing pathological changes, but also for an estimation of efficiency of treatment, in particular antineoplastic therapy, and also for definition of relapses and accompanying complications. Diagnostics by means of CТ is based on direct radiological symptoms, i.e. definition of exact localization, the form, the sizes of separate bodies and the pathological center, and, that is especially essential, on parameters of density or absorption.

Considering statement of the given problem there was a necessity of engineering a system of automation of the processing received at inspection, carried out on the basis of the COMPUTER. Process of manual processing of tomograms now is the big problem because, it leads to big expenses of time and accordingly reduces throughput of cabinets of a computer tomography.

Realization of the given system is carried out by means of Delphi 7 with the purpose of processing of images, and also for realization of the interface convenient for the medical personnel. The given program allows to make processing of computer tomograms and to diagnose.

THE PURPOSES AND PROBLEMS OF WORK

In master's thesis I plan to create Special Computer Sistem (SCS) that realize 2 problems:

1. Processing tomograms of new growths of a liver.
2. Statement of the diagnosis on the basis of results of researches of new growths of a liver.

The program should carry out a filtration of the image, regulation of brightness and contrast, and also edge-point linking and allocation of objects. Calculation of parameters of objects (their sizes, the area, number of objects, their brightness) will be made further and to assume, whether are these objects new growths. For definition of type of new growths, it is necessary to investigate what brightness have new growths on the tomogram. In this question the help of doctors and the engineers working with computer tomographs is required.

Statement of the diagnosis will be carried out by results of researches and analyses. For statement of the diagnosis Hamming and Hopfield neural networks will be used . In the further it is necessary to find out, what researches to use at diagnosing. Consultation of doctors - oncologists is necessary for the decision of this question .

THE REVIEW OF EXISTING METHODS OF PROCESSING OF TOMOGRAMS

The given direction of researches is rather new and up to the end not developed. In existing software products, processing of the image is limited to a filtration, regulation of brightness and contrast only. Thus the given software products do not carry out edge-point linking and allocation of objects.А doctor is compelled to carry out this all manually, that entails increase in time of processing of tomograms. I shall try to solve this problem due to creation of the software product capable by means of mathematical methods to allocate on an available tomogram objects, to define their parameters (the sizes, the area, number of objects, their brightness) and to assume, whether are these objects new growths.

CONCLUSIONS

In the work I have tried to choose the optimal methods of contrast enhancement and filtrations. Using methods of binarization and edge-point linking, I allocate objects on the tomogram, I define their sizes and the area. On the basis of the given approach, the computer system based on neuronet modelling is developed. The neural network capable adequately to react to entrance influences and to provide statement of the diagnosis of new growths of a liver that proves correctness of the chosen method. The realizable software product will comprise experience and knowledge of leading experts in this area. Thus, use of neural networks at processing computer tomograms of new growths of a liver will allow to create effective SCS. Program which is doing during master’s scientific work is unique as it has no analogues nowadays.