{"id":211,"date":"2015-06-18T06:40:21","date_gmt":"2015-06-18T06:40:21","guid":{"rendered":"http:\/\/biomedpharmajournal.org\/?p=211"},"modified":"2020-04-25T02:49:26","modified_gmt":"2020-04-25T02:49:26","slug":"measurement-of-lower-and-upper-imt-from-ultrasound-video-frames","status":"publish","type":"post","link":"https:\/\/biomedpharmajournal.org\/staging\/vol8no1\/measurement-of-lower-and-upper-imt-from-ultrasound-video-frames\/","title":{"rendered":"Measurement of Lower and Upper IMT from Ultrasound Video Frames"},"content":{"rendered":"

Introduction<\/strong><\/p>\n

Ultrasound scanning exposes the inspected part of the body to sound waves which have high frequency as well as it shows the movement and shape of internal parts of the body organ. It is a non \u2013invasive test. Doppler ultrasound is also used to diagnose carotid arteries. Ultra sound scanning modes are A-mode, B-mode and M-mode. A-mode scanning represents the amplitude mode here this strength of echo signal is measured and displayed in directional continuous signal. It is used for ophthalmology. B-mode scanning represents the brightness mode; here the strength of the echo signal is viewed as 2D gray scale image. M- mode scanning refers to A-mode with time and used for cardiology. Doppler carotid ultrasound Image is used for radiologist to evaluate tumors blood flow in blood vessels, abnormalities in lymph nodes and lymphatic vessels. Atherosclerosis is a disease which due to accumulation of cholesterol, protein and lipid arterial wall causes thickening. Thickening leads to narrowing of vessels (caused by plaque) and blocks the blood flow (clots).<\/p>\n

Blockage or narrowing carotid arteries increase the risk of having a heart disease or stroke. Measurement of 2 innermost layers of arterial wall is called IMT otherwise measurement of Atherosclerosis, Which gives the distance of borders between lumen-intima and media-adventitia.<\/p>\n

Many works have been carried out for the measurement of IMT like:VanderMeer et al [1] proposed that thickening of arterial wall layers like intima and medial leads to cardiovascular disease. Bianchini et al [2] present a method for arterial stiffness evaluation. By measuring the thickness change in heart with local pulse pressure and compared that value with femoral pulse wave velocity for arterial stiffness evaluation.Destrempes et al.[3] proposed a model based on properties of carotid arteries tissue like elastogram are modelled by Nakagami distribution and Expectation Maximization algorithm is used to compute the posterior estimation of it. An automatic system was developed by Linguri et al [4] which depends on a analytical approach for measurement of carotid intima media thickness and to determine the intrinsic accuracy of the system. Improved active contour technique by a multiresolution analysis is presented for measuring lumen diameter and IMT [5]. This automatic technique and manual segmentation technique for blood vessel are compared based on the parameters like co-efficient of variability and correlation. This improved active contour technique produces a best result. In this paper IMT measurement is based on unsupervised computer aided detection (CAD) algorithm [6].<\/p>\n

The Intima Media Complex (IMC)[7] is best visualized in longitudinal sections of the common carotid artery. It is composed of two quasi-parallel lines that represent the Lumen Intima (LI) and the media adventitia (MA) interfaces. The thickening of the intima and medial layers is caused by inflammatory-fibro proliferative responses to various forms of insult. It involves lipid accumulation and the migration and proliferation of many cells in the subintimal and medial layers, which result in the formation of plaques. It is the rupture of such plaques that causes myocardial infarcts (heart attacks), cerebrovascular events (strokes), peripheral vascular disease (gangrene), and kidney infarcts. The Rotterdam study [8] indicates that the risk of myocardial infarction increases 43% per standard deviation increase (0.163 mm) in common carotid IMT [8].<\/p>\n

Atherosclerotic disease occurs when IMT is higher than 0.9 to 1.0 mm. indicates a potential atherosclerotic disease. Hence, the robust segmentation and measurement of the IMT has a considerable impact in the early diagnosis of atherosclerosis, prognosis prediction, and in the monitoring of responses to lifestyle and prescribed pharmacological treatments. The accuracy of the manual tracing process was analyzed in \u201cAssociations of edge-detected and manual-traced common carotid Intima-media thickness measurements [9]. Disadvantages of the existing system are that they do not take into consideration the speckle noise or the image normalization in the ultrasound image. They require manual correction after automatic tracing. This increases the computational time and the error probability associated with the identification of the IMC in video data. In the existing we had used watershed segmentation algorithm.<\/p>\n

Materials and Methods<\/strong><\/p>\n

In this paper, an algorithm to segment the carotid artery in the ultrasound video or in the first frame of the video sequence is developed completely by unsupervised and includes the following computational steps: automatic Region of Interest (ROI) detection by using adaptive thresholding, and binarization by using Otsu\u2019s thresholding. The segmented carotid artery region consists of multiple noises. Among them the content of the speckle noise will be high. So in order to remove the noise we can use filters like lee, kuan, wavelet denoising and SRAD (Speckle Reduction Anisotrophic Diffusion).<\/p>\n

This process is Despeckling. After applying binarization technique and by performing morphological operations the carotid artery region in the image will be in enhanced form, such as it is easy for us to measure the Intima Media Thickness (IMT). Finally the Upper and Lower Intima Media Thickness range will be measured. Figure 2.1 shows the overall proposed block diagram.<\/p>\n\n\n\n
\"Figure<\/td>\nFigure 1 : Overall Proposed Block Diagram<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Image Acquisition<\/strong><\/p>\n

Image acquisition process generates large ultrasound image sequences that are interpreted using either manual annotation procedures or histological validated semi-automatic image-processing environments. The Ultrasound image sequences of longitudinal sections of the Common Carotid Artery (CCA) of some patients are captured. The captured image which is in color format is converted into gray<\/p>\n

Scale image. Figure 2.2 shows the ultrasound frame in color image and Figure 2.3 shows the converted gray scale image.<\/p>\n\n\n\n
\"Figure<\/td>\nFigure 2 : Input Frame<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n\n\n\n
\"Figure<\/td>\nFigure 3 : Converted Grayscale Frame<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

ROI Extraction<\/strong><\/p>\n

Region of interest (ROI) detection is one of the steps used in segmenting the Intima Media Complex (IMC) in the image sequence which is completely unsupervised. The main principle behind the proposed approach is to identify the location of the far wall interface using a suite of image processing steps that combine the information contained in the intensity domain with knowledge relating to the anatomical structure of the carotid artery. From the input image, the required portion is cropped and that cropped image is used for further processing. Figure 2.4 shows the ROI selected from the input image. Figure 2.5 and Figure 2.6 shows the cropped upper and lower part for the measurement of intima media thickness respectively.<\/p>\n

 <\/p>\n\n\n\n
\"Figure<\/td>\nFigure 4 : ROI for measurment of upper and lower intima thickness<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n\n\n\n
\"Figure<\/td>\nFigure 5 : Cropped Upper Image<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n\n\n\n
\"Figure<\/td>\nFigure 6 : Lower cropped image<\/strong><\/p>\n

Click here to view figure<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n

 <\/p>\n

Despeckling\/Preprocessing<\/strong><\/p>\n

Speckle in ultrasound scan are granular in structure which occurs due to coherent interferences of back scattered echos from the scatters that are smaller than spatial resolution of ultrasound system. It is a multiplicative noise. Speckle pattern depends on many parameters of image and tissues in medical image. Speckle affects the contrast and resolution of the image. Speckle pattern are classified into fully formed speckle, Non randomly distributed with long range order and Non randomly distributed with short range order classes.<\/p>\n

Main purpose in despeckling are :<\/p>\n