BIOGRAPHY
ABSTRACT (RUS.)
|
ABSTRACT
Introduction.
Malignant new growths are one of the most dangerous medical and biologic and social and economic problems. Desease and death rate from a cancer constantly grow in connection with an adverse ecological situation and significant ageing of the population.
In 2000 in Ukraine it is registered 156,6 thousand primary oncological patients. According to with estimations, their total will make up to 2010 180 thousand, till 2020 - from above 200 thousand persons. Proof tendencies to growth of desease and death rate from a cancer, an opportunity of oncological effects as a result of failure on the Chernobyl atomic power station testify to an extreme acuteness of a problem.
At the present stage of development, in connection with improvement of computer technologies, wide application has received use of modern computer systems (CS) in technics, aircraft, bank sphere, medicine... Especial value has use CS in medicine. CS with the big share of reliability allow to spend diagnosing of illness at an early stage of its development, helping doctors to make the preliminary diagnosis.
The basic problem at creation of the automated system of diagnostics malignant new growths of a brain – is search of effective, high-speed and reliable methods of processing of images of the brain received from a computer tomograph.
Despite of presence of software products and experimental systems on processing medical images, the circle of problems solved in them is rather limited and does not allow the doctor, having allocated a site of a brain with suspicion on malignant new growths, automatically to receive acknowledgement or a refutation of the suspicion, in fact except for obviously expressed tumour on the image there can be indistinct, washed away, or too fine educations (structure) which demand for diagnostics of special attention.
Therefore development such CS which can help doctors with fast statement of the maximal exact diagnosis, reducing thus time, is the important and actual problem.
The purposes and problems.
The purpose of the given job is development of the automated subsystem of diagnosing of pathologies of development of a brain on the pictures received with computer tomograph.
For achievement of an object in view it is necessary to solve following problems:
To collect and analyse existing methods for processing pictures of a brain;
Development of statement of a problem of definition of the pathological center on image a brain;
Choice of a method for definition and processings of the centers which are classified as pathological;
Realization of the chosen methods on the computer.
Search of algorithms was made for processing tomograms among methods of comparison with the standard. Thus attributes of the reference image and attributes of the entered images are considered from the point of view of the textural analysis since the structure is the important characteristic which is present at all images, since the images received by means of plane and satellite devices and finishing microscopic images in biomedical researches.
Object of research.
Object of research are it is black – white images of the brain, received by means of CÒ. The Object of research are algorithms of segmentation, a vectoring and allocation of objects on medical images.
Scientific novelty.
Use of the algorithms offered in job at processing computer tomograms of new growths of a brain will allow to create effective CS. The core novelty jobs is use for an estimation of the allocated center of a brain simultaneously 2 approaches: the textural and planimetric analysis. Comparison of the received results will allow to draw a conclusion on a status of the chosen site and to estimate accuracy and speed of each of approaches.
© DonNTU 2008 Krivopysk T.I.
|
