NOISE REDUCTION OF LUNGS COMPUTED TOMOGRAPHY IMAGES

Abstract

This article deals with the problem of improving the quality of the decision support system by improving the preprocessing of human lungs computed tomography images. These images are low-contrast and contain various noise, so pre-processing such images is an actual current issue. This problem is very important because quality of the preprocessing determines the efficiency of the decision support system. decision support system is used in the processing of computerized tomograms to reduce the level of noise in images in order to increase the probability of correct diagnosis. The aim of the work was to create a better method for apply in this applied area. This article has been analyzed existing methods of reducing the level of noise in the images of humans lungs computed tomography. This work revealed the main disadvantages of the existing methods. Combined method with contrasting was proposed for reducing the level of noise on images of computed tomography, on the basis of the received information. The efficiency of the proposed method was evaluated by three indicators and compared with the efficiency of the considered methods. Based on the data of these assessments, the method with contrasting turned out to be the best among those considered, which shows the expediency of applying the proposed method in real decision support systems in medicine. The application of the combined method with contrasting which proposed in the article for the preprocessing of images of humans lungs computed tomography will improve the quality of the system as a whole, which in turn will increase the probability of correct diagnosis..