Fig 1 - uploaded by Naser Movahhedinia
Content may be subject to copyright.
Source publication
This paper presents a fully automated approach to detect the intima and media-adventitia borders in intravascular ultrasound images based on parametric active contour models. To detect the intima border, we compute a new image feature applying a combination of short-term autocorrelations calculated for the contour pixels. These feature values are e...
Contexts in source publication
Context 1
... w(l) is a small window of length L. In the proposed method, STA is computed from R l (-L/2) to R l (L/2) around each index t of the signal f(t). Then, the sum of the obtained STA values is calculated and placed in location t as shown in Fig. 1. The result is a new function (signal) F R with the same length as the original signal f produced by the sum- mation of STA in a small window around the index. We name this new function as cumulative short-term autocor- relation (CSTA). CSTA can be generalized to 2-D image signals defined at horizontal lag k 1 and vertical lag k 2 for the image I ...
Context 2
... IVUS images with different scenarios, corre- sponding manual tracing (AE1E2) and automatic segmen- tation results using the proposed method, have been presented in Fig. 5, column 1-column 3 respectively. Images without large calcified plaques and side branches are shown in Fig. 5 (Row 1, 2). To indicate the strength of the proposed methods rather than previous models, we compared the proposed active contours (PAC) with the traditional active contour (TAC) model based on Kass energy function and the gradient vector flow (GVF) model that have been widely used in previous works for border detection in IVUS images [15,16,23,24]. Figure 6 shows the results of detecting intima borders for three models and manual tracing in two IVUS images. The results of the TAC model are displayed in Fig. 6(row 1). As seen in this figure, the TAC model could not identify the intima borders in both images caused by speckle noise in the lumen area. Second row of Fig. 6 indicates the results of the GVF model. When noise and clutters in the lumen area were low, this model was able to find the intima border (Fig. 6(row 2(a))). However, when there were heavy noise and clutter patterns in the lumen, the GVF model was also involved with difficulty and could not truly find the intima (Fig. 6(row 2(b))). Detected intima borders by the PAC model are shown in Fig. 6(row 3). This model using the NCSTA feature values, could correctly extract intima even though the lumen region is too clut- tered and noisy (Fig. 6(row 3(b))). The segmentation results of media-adventitia boundaries obtained by three models and manual tracing for two other IVUS images are shown in Fig. 7. Detected media-adventitia boundaries by the TAC model are shown in Fig. 7(row 1). As shown in this figure, owning to presence of calcified plaques, this model could not truly detect the media-adventitia bound- aries in both images. The results obtained by the GVF model are displayed in Fig. 7(row 2). When there were not large calcifications, the GVF model could identify the media-adventitia boundary (Fig. 7(row 2(a))). But, due to existing of large calcified plaques with strong edges in Fig. 7(row 2(b)); the GVF model could not correctly find the media-adventitia boundary in this image. Contrary to both TAC and GVF models, the PAC model based on the segmentation algorithm described in ''Details of segmen- tation algorithm'' section could exactly detect the media- adventitia boundaries in both images ( Fig. 7(row ...
Context 3
... IVUS images with different scenarios, corre- sponding manual tracing (AE1E2) and automatic segmen- tation results using the proposed method, have been presented in Fig. 5, column 1-column 3 respectively. Images without large calcified plaques and side branches are shown in Fig. 5 (Row 1, 2). To indicate the strength of the proposed methods rather than previous models, we compared the proposed active contours (PAC) with the traditional active contour (TAC) model based on Kass energy function and the gradient vector flow (GVF) model that have been widely used in previous works for border detection in IVUS images [15,16,23,24]. Figure 6 shows the results of detecting intima borders for three models and manual tracing in two IVUS images. The results of the TAC model are displayed in Fig. 6(row 1). As seen in this figure, the TAC model could not identify the intima borders in both images caused by speckle noise in the lumen area. Second row of Fig. 6 indicates the results of the GVF model. When noise and clutters in the lumen area were low, this model was able to find the intima border (Fig. 6(row 2(a))). However, when there were heavy noise and clutter patterns in the lumen, the GVF model was also involved with difficulty and could not truly find the intima (Fig. 6(row 2(b))). Detected intima borders by the PAC model are shown in Fig. 6(row 3). This model using the NCSTA feature values, could correctly extract intima even though the lumen region is too clut- tered and noisy (Fig. 6(row 3(b))). The segmentation results of media-adventitia boundaries obtained by three models and manual tracing for two other IVUS images are shown in Fig. 7. Detected media-adventitia boundaries by the TAC model are shown in Fig. 7(row 1). As shown in this figure, owning to presence of calcified plaques, this model could not truly detect the media-adventitia bound- aries in both images. The results obtained by the GVF model are displayed in Fig. 7(row 2). When there were not large calcifications, the GVF model could identify the media-adventitia boundary (Fig. 7(row 2(a))). But, due to existing of large calcified plaques with strong edges in Fig. 7(row 2(b)); the GVF model could not correctly find the media-adventitia boundary in this image. Contrary to both TAC and GVF models, the PAC model based on the segmentation algorithm described in ''Details of segmen- tation algorithm'' section could exactly detect the media- adventitia boundaries in both images ( Fig. 7(row ...
Similar publications
Intravascular ultrasound (IVUS) is a catheter-based medical imaging technique that produces cross-sectional images of blood vessels. In this paper, we present a method for the segmentation of the luminal border using IVUS radio frequency (RF) data. Specifically, we parameterize the lumen contour using Fourier series. This contour is deformed by min...
Citations
... The algorithm clustered the images by k-means algorithm, and determined the lumen region by using the average circularity, and then it used the initial contour as the input of the model to extract the final contour. Another algorithm based on active contour model is Vard method [17]. For the intima, a new image feature was calculated by short-term autocorrelation combination of contour pixels to define the energy function of the model. ...
The extraction of intima and adventitia from intravascular ultrasound (IVUS) images is of great significance for the diagnosis and treatment of coronary artery disease. However, traditional IVUS image segmentation methods have complex modeling, poor robustness, and need to design different algorithms for internal and external boundaries. This article proposes a method of simultaneous intima and adventitia. We firstly detect extreme regions of the pre-processed image, and then design a screening vector to extract two extreme regions representing the lumen and media. After that, the convex–concave boundaries of the two regions are adjusted by the opening operation with a variable radius of the structural element and the superposition of a circle. Finally, the contours are fitted with ellipse to complete the segmentation. In order to evaluate the performance of the method, we first qualitatively display the extreme region contours and the final contours, and compare the results with those drawn manually by clinical experts. Then, we use Jaccard measure, Dice coefficient, percentage of area difference and Hausdorff distance to test the robust performance and generalization performance, and the index values of inner and outer borders are 0.92 ± 0.03, 0.96 ± 0.02, 0.06 ± 0.04, 0.26 ± 0.07 mm, and 0.92 ± 0.04, 0.95 ± 0.02, 0.06 ± 0.05, 0.29 ± 0.08 mm, respectively. Besides, we make a quantitative comparison with the relevant studies. Experiment results show that the proposed automatic methods not only have high accuracy, but also have good robustness.
... Wang et al. (2014) also introduced a new denoising approach based on framelet regularization. An automated approach for segmentation of intravascular ultrasound images is also studied in Vard et al. (2012). ...
... Wang et al. (2014) also introduced a new denoising approach based on framelet regularization. An automated approach for segmentation of intravascular ultrasound images is also studied in Vard et al. (2012). ...
Medical Ultrasound is a diagnostic imaging technique based on the application of ultrasound in various branches of medical sciences. It can facilitate the observation of structures of internal body, such as tendons, muscles, vessels and internal organs such as male and female reproductive system. However, these images usually degrade by a special kind of multiplicative noise called speckle. The main effects of speckle noise in the ultrasound images appear in the edges and fine details which lead to reduce their resolution and consequently make difficulties in medical diagnosing. Therefore, reducing of speckle noise seriously plays an important role in image diagnosing. Among the various methods that have been proposed to reduce the speckle noise, there exists a class of approaches that firstly convert multiplicative speckle noise into additive noise via log-transform and secondly perform the despeckling process via a directional filter. Usually, the additive noises are mutually uncorrelated and obey a Gaussian distribution. On the other hand, non-subsampled shearlet transform (NSST), as a multi scale method, is one of the effective methods in image processing, specially, denoising. Since NSST is shift invariant, it diminishes the effect of pseudo-Gibbs phenomena in the denoising. In this paper, we describe a simple image despeckling algorithm which combines the log-transform as a pre-processing step with the non-subsampled shearlet transform for strong numerical and visual performance on a broad class of images. To illustrate the efficiency of the proposed approach, it is applied on a sample image and two real ultrasound images. Numerical results illustrate that the proposed approach can obtain better performance in term of peak signal to noise ratio (PSNR) and structural similarity (SSIM) index rather than existing state-of-the-art methods.
... In this regard, firstly, the internal and external borders of vessel before extracting calcifications should be identified. We utilized the method presented by Vard et al... [39] to detect the lumen and media-adventitia borders of the coronary arteries. Briefly, in this approach, the IVUS image is converted from the Cartesian coordinates to the polar coordinates. ...
This paper presents a novel combined energy functional based on edge and region information for active contour model, which can be applied to segment textured images containing low contrast and high illumination variations. In the proposed method, the edge features are calculated based on the phase-based approaches derived from the monogenic signal, which are robust to illumination variations in the image. These feature values are used in an edge energy functional to assist the active contour evolving towards the true object boundaries. To extract the region features, at first, we compute the normalized accumulated short-term autocorrelation (NASTA) values for the image, which suppress background clutter and enhance dissimilarities between objects and background. Next, a local cumulative distribution function (LCDF) of NASTA is calculated for every pixel in a local window around it and is used as a region feature for that pixel. Then, the obtained features are employed to define a new localized region-based energy functional that can correctly segment regions with same intensity mean and variance. The proposed edge and region energy terms are integrated and a regularization term is added to them to form our combined active contour. Experimental results indicate remarkable advantages of proposed method comparing to existing combination models.
... Contour-based evaluation metrics, which included Mean Distance (MD), Hausdorff Distance (HD) [32] and Relative Distance Error (RDE) [33], were calculated by comparing the estimated contour by the algorithm with the standard results manually traced by two experts (T1 and T2) and the resultant average of T1 and T2 (T1T2). The MD and HD describe the average and maximum point-to-point Euclidian distance between estimated and manual contours respectively. ...
Estimation of missing digital information is mostly addressed by one or two-dimensional signal processing methods; however, this problem can emerge in multi-dimensional data including 3D images. Examples of 3D images dealing with missing edge information are often found using dental micro-CT, where the natural contours of dental enamel and dentine are partially dissolved or lost by caries. In this paper, we present a novel sequential approach to estimate the missing surface of an object. First, an initial correct contour is determined interactively or automatically, for the starting slice. This contour information defines the local search area and provides the overall estimation pattern for the edge candidates in the next slice. The search for edge candidates in the next slice is performed in the perpendicular direction to the obtained initial edge in order to find and label the corrupted edge candidates. Subsequently, the location information of both initial and nominated edge candidates are transformed and segregated into two independent signals (X-coordinates and Y-coordinates) and the problem is changed into error concealment. In the next step, the missing samples of these signals are estimated using a modified Tikhonov regularization model with two new terms. One term contributes in the denoising of the corrupted signal by defining an estimation model for a group of mildly destructed samples, and the other term contributes in the estimation of the missing samples with the highest similarity to the samples of the obtained signals from the previous slice. Finally, the reconstructed signals are transformed inversely to edge pixel representation. The estimated edges in each slice are considered as initial edge information for the next slice and this procedure is repeated slice by slice until the entire contour of the destructed surface is estimated. The visual results as well as quantitative results (using both contour-based and area-based metrics) for seven image datasets of tooth samples with considerable destruction of the dentin-enamel junction (DEJ) demonstrates that the proposed method can accurately interpolate the shape and the position of the missing surfaces in computed tomography images in both two and three dimensions (e.g. 14.87 ±3.87 μ m of mean distance (MD) error for the proposed method versus 7.33 ±0.27 μm of MD error between human experts and 1.25 ±~0 % error rate (ER) of the proposed method versus 0.64 ±~0 % of ER between human experts (~1% difference)).
... Over the past few years, intravascular ultrasound (IVUS) has become very useful for studying atherosclerotic disease. IVUS is a catheter-based imaging technique providing realtime high-resolution images of the vessel wall and lumen [1,2]. ...
Intravascular ultrasound (IVUS) is clinically available for visualizing coronary arteries. However, it suffers from acoustic shadow areas and ring-down artifacts as two of the common issues in IVUS images. This paper introduces an approach which can overcome these limitations. As shadow areas were displayed behind hard plaques in the IVUS grayscale images, calcified plaques were first segmented by using Otsu threshold. Then, active contour, histogram matching, and local histogram matching are implemented. In addition, a new modified circle Hough transform is introduced to remove the ring-down artifacts from IVUS images. In order to evaluate the efficacy of this new method in detection of shadow and ring-down regions, 300 IVUS images are considered. Sensitivity of 89% and specificity of 92% are achieved from a comparison in revelation of calcium along with shadow in the proposed method and virtual histology images. Also, area differences of \(5.83 \pm 3.3\) and \(5.65 \pm 2.83\) are obtained, respectively, for ring-down and shadow domain when compared to measures performed manually by a clinical expert.
... A parametric active contour model for the segmentation of lumen and media-adventitia contours in IVUS images of coronary arteries was proposed by Vard et al. (2012). For the lumen contour segmentation, a 85 method based on the short-term autocorrelation (STA) is used to remove speckle noise from the lumen region. ...
Image assessment of the arterial system plays an important role in the diagnosis of cardiovascular diseases. The segmentation of the lumen and media-adventitia in intravascular (IVUS) images of the coronary artery is the first step towards the evaluation of the morphology of the vessel under analysis and the identification of possible atherosclerotic lesions. In this study, a fully automatic segmentation of the lumen in IVUS images of the coronary artery is presented. The proposed method relies on the K-means algorithm and the mean roundness to identify the region corresponding to the potential lumen. An approach to identify and eliminate side branches on bifurcations is also proposed to delimit the area with the potential lumen regions. Additionally, an active contour model is applied to refine the contour of the lumen region. In order to evaluate the segmentation accuracy, the results of the proposed method were compared against manual delineations made by two experts in 326 IVUS images of the coronary artery. The average values of the Jaccard measure, Hausdorff distance, percentage of area difference and Dice coefficient were 0.88 ± 0.06, 0.29 ± 0.17 mm, 0.09 ± 0.07 and 0.94 ± 0.04, respectively, in 324 IVUS images successfully segmented. Additionally, a comparison with the studies found in the literature showed that the proposed method is slight better than the majority of the related methods that have been proposed. The new automatic segmentation method is shown to be effective in detecting the lumen in IVUS images without using complex solutions and user interaction.
... Consequently, as can be seen from In addition, we performed the linear regression analyses [30,31] as a powerful technique to consider correlation of our method and ground truth. For this purpose, in each of the 45 images, we calculated the area of WBCs and nuclei that were obtained by using the proposed method Figure 6: From left to right: original images, visual results of manual segmentation by the expert, results of automatic segmentation by proposed method, and results of automatic segmentation by the algorithm of Saeedizadeh et al. [24] Ghane, et al.: Automatic segmentation of white blood cells from microscopic images and determined by an expert. ...
Recognition of white blood cells (WBCs) is the first step to diagnose some particular diseases such as acquired immune deficiency syndrome, leukemia, and other blood-related diseases that are usually done by pathologists using an optical microscope. This process is time-consuming, extremely tedious, and expensive and needs experienced experts in this field. Thus, a computer-aided diagnosis system that assists pathologists in the diagnostic process can be so effective. Segmentation of WBCs is usually a first step in developing a computer-aided diagnosis system. The main purpose of this paper is to segment WBCs from microscopic images. For this purpose, we present a novel combination of thresholding, k-means clustering, and modified watershed algorithms in three stages including (1) segmentation of WBCs from a microscopic image, (2) extraction of nuclei from cell’s image, and (3) separation of overlapping cells and nuclei. The evaluation results of the proposed method show that similarity measures, precision, and sensitivity respectively were 92.07, 96.07, and 94.30% for nucleus segmentation and 92.93, 97.41, and 93.78% for cell segmentation. In addition, statistical analysis presents high similarity between manual segmentation and the results obtained by the proposed method.
... Consequently, as can be seen from In addition, we performed the linear regression analyses [30,31] as a powerful technique to consider correlation of our method and ground truth. For this purpose, in each of the 45 images, we calculated the area of WBCs and nuclei that were obtained by using the proposed method Figure 6: From left to right: original images, visual results of manual segmentation by the expert, results of automatic segmentation by proposed method, and results of automatic segmentation by the algorithm of Saeedizadeh et al. [24] Ghane, et al.: Automatic segmentation of white blood cells from microscopic images and determined by an expert. ...
Recognition of white blood cells (WBCs) is the first step to diagnose some particular diseases such as acquired immune deficiency syndrome, leukemia, and other blood-related diseases that are usually done by pathologists using an optical microscope. This process is time-consuming, extremely tedious, and expensive and needs experienced experts in this field. Thus, a computer-aided diagnosis system that assists pathologists in the diagnostic process can be so effective. Segmentation of WBCs is usually a first step in developing a computer-aided diagnosis system. The main purpose of this paper is to segment WBCs from microscopic images. For this purpose, we present a novel combination of thresholding, k-means clustering, and modified watershed algorithms in three stages including (1) segmentation of WBCs from a microscopic image, (2) extraction of nuclei from cell's image, and (3) separation of overlapping cells and nuclei. The evaluation results of the proposed method show that similarity measures, precision, and sensitivity respectively were 92.07, 96.07, and 94.30% for nucleus segmentation and 92.93, 97.41, and 93.78% for cell segmentation. In addition, statistical analysis presents high similarity between manual segmentation and the results obtained by the proposed method.
... According to the energy, you will discover two main families of active contour models: edge-based models (Pratondo, Chui & Ong, 2016) and region-based models (Vard, Jamshidi & Movahhedinia, 2012). Edge-based productive contour models count on the graphic gradient to halt the growing contours about the desired subject boundaries. ...
Medical applications became a boon to the healthcare industry. It needs correct and fast segmentation associated with medical images for correct diagnosis. This assures high quality segmentation of medical images victimization. The Level Set Method (LSM) is a capable technique, however the quick process using correct segments remains difficult. The region based models like Active Contours, Globally Optimal Geodesic Active Contours (GOGAC) performs inadequately for intensity irregularity images. During this cardstock, we have a new tendency to propose an improved region based level set model motivated by the geodesic active contour models as well as the Mumford-Shah model. So that you can eliminate the re-initialization process of ancient level set model and removes the will need of computationally high priced re-initialization. Compared using ancient models, our model are sturdier against images using weak edge and intensity irregularity.
