Automatic segmentation of intravascular ultrasound images: A texture-based approach

Mojsilović, Aleksandra; Popović, M.; Amodaj, Nenad; Babic, Rade; Ostojić, Miodrag

doi:10.1007/bf02648131

Cited by 28 publications

(37 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(4) The snake is implemented as a close contour [8,36,40] that might not be that suitable for the IMC segmentation. (5) They require manual correction after automatic tracing [6,23,24,34]. (6) In a number of cases there was no ground truth segmentation delineations from experts to compare to the computer assisted methods [6,24].…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, the observation of the IMT becomes more difficult as the age of patients increases, due to the presence of acoustic holes (echo-dropouts) in the adventitia layer [15,35]. The intimal band may appear as a thin low contrast structure and therefore it is difficult to reliably draw boundaries because smoothing can move the structure edges or make them disappear [24]. Traditionally, the IMT is measured by manual delineation of the intima and the adventitia layer [17,34,35].…”

Section: Introductionmentioning

confidence: 99%

“…The problems that are associated with the computerassisted border tracing segmentation procedures (also presented in Table 1) are the following: (1) They do not take into consideration the speckle noise or the image normalization in the ultrasound image [24]. (2) They are sensitive to the initial snake contour [6], or initial seed points, which need to be placed manually Fig.…”

Section: Introductionmentioning

confidence: 99%

“…If the initial contour is placed far away from the boundary of interest then the snake will not be attracted [6]. (3) Some weighting factors that should be tuned due to the varied characteristics of the ultrasound instrumentation must be entered manually or empirically [24,34,37,40]. Some other weights may be adjusted by a training procedure, which might be long and requires experts tracing [17,23].…”

Section: Introductionmentioning

confidence: 99%

“…(5) They require manual correction after automatic tracing [6,23,24,34]. (6) In a number of cases there was no ground truth segmentation delineations from experts to compare to the computer assisted methods [6,24]. (7) Different measurement procedures were used between the manual and the snakes segmentation methodologies (see Fig.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Snakes based segmentation of the common carotid artery intima media

et al. 2007

View full text Add to dashboard Cite

Ultrasound measurements of the human carotid artery walls are conventionally obtained by manually tracing interfaces between tissue layers. In this study we present a snakes segmentation technique for detecting the intima-media layer of the far wall of the common carotid artery (CCA) in longitudinal ultrasound images, by applying snakes, after normalization, speckle reduction, and normalization and speckle reduction. The proposed technique utilizes an improved snake initialization method, and an improved validation of the segmentation method. We have tested and clinically validated the segmentation technique on 100 longitudinal ultrasound images of the carotid artery based on manual measurements by two vascular experts, and a set of different evaluation criteria based on statistical measures and univariate statistical analysis. The results showed that there was no significant difference between all the snakes segmentation measurements and the manual measurements. For the normalized despeckled images, better snakes segmentation results with an intra-observer error of 0.08, a coefficient of variation of 12.5%, best Bland-Altman plot with smaller differences between experts (0.01, 0.09 for Expert1 and Expert 2, respectively), and a Hausdorff distance of 5.2, were obtained. Therefore, the pre-processing of ultrasound images of the carotid artery with normalization and speckle reduction, followed by the snakes segmentation algorithm can be used successfully in the measurement of IMT complementing the manual measurements. The present results are an expansion of data published earlier as an extended abstract in IFMBE Proceedings (Loizou et al. IEEE Int X Mediterr Conf Medicon Med Biol Eng POS-03 499:1-4, 2004).

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Snakes based segmentation of the common carotid artery intima media

et al. 2007

View full text Add to dashboard Cite

show abstract

Automatic quantification of histological studies in allergic asthma

et al. 2008

View full text Add to dashboard Cite

The evaluation of new therapies to treat allergic asthma makes frequent use of histological studies. Some of them are based on microscope observation of stained paraffin lung sections to quantify cellular infiltrate, an effect directly related to allergic processes. Currently, there is no software tool available for doing this quantification automatically. This paper presents a methodology and a software tool for the quantification of cellular infiltrate in lung tissue images in an allergic asthma mouse model. The image is divided into regions of equal size, which are then classified by means of a segmentation algorithm based on texture analysis. The classification uses three discriminant functions, built from parameters derived from the histogram and the co-occurrence matrix. These functions were calculated by means of a stepwise discriminant analysis on 79 samples from a training set. Results provided a correct classification of 96.8% on an independent test set of 251 samples labeled manually. Regression analysis showed a good agreement between automatic and manual methods. A reliable and easy to implement method has been developed to provide an automatic method for quantifying microscopy images of lung histological studies. Results showed similar accuracy to that provided by an expert, while allowing analyzing a much larger number of fields in a repea- ASTHMA is a chronic inflammatory disease of the lung, characterized by airwayhyperresponsiveness to a variety of stimuli, eosinophilic inflammation of the airways, mucus hypersecretion, and elevated serum IgE levels (1). Mortality of asthma has increased worldwide, despite the use of currently available medications, underlining the need for the development of novel therapies (2-4).Previous studies indicate that murine models are useful for studying allergic diseases, including certain aspects of bronchial asthma such as cellular tissue inflammation and pulmonary function. Three studies are commonly performed in this type of experiments to assess the effects: immunologic parameters (immunoglobulines and cytokines), pulmonary function (bronchial hyperactivity), and histological studies (cellular infiltrate and bronchial mucus secretion). The latter are currently assessed by visual inspection by an expert, as there is no automatic analysis tool available. Such a tool would speed up the process, also providing a more repeatable quantification.In theory, the amount of cellular infiltrate could be accurately assessed by means of an appropriate segmentation algorithm (i.e., border detection) identifying each cell nucleus on the image. The main difficulty for this individual cell segmentation arises from the existence of cell aggregates, which hinders the detection of the contours. This is a difficult task for most image-processing algorithms and the authors do not know of any previous successful attempt. Furthermore, the nuclei of cells in the bronchial wall are similarly stained and have the same size, making it difficult to

show abstract

Optoacoustic sensing of hematocrit to improve the accuracy of hybrid fluorescence‐ultrasound intravascular imaging

Bozhko

Karlas

Gorpas

et al. 2018

Journal of Biophotonics

View full text Add to dashboard Cite

Hybrid intravascular fluorescence-ultrasound imaging is emerging for reading anatomical and biological information in vivo. By operating through blood, intravascular near-infrared fluorescence (NIRF) detection is affected by hemoglobin attenuation. Improved quantification has been demonstrated with methods that correct for the attenuation of the optical signal as it propagates through blood. These methods assume an attenuation coefficient for blood and measure the distance between detector and the vessel wall by observing the intravascular ultrasound images. Assumptions behind the attenuation employed in correction models may reduce the accuracy of these methods. Herein, we explore a novel approach to dynamically estimate optical absorption by using optoacoustic (photoacoustic) measurements. Adaptive correction is based on a trimodal intravascular catheter that integrates fluorescence, ultrasound and optoacoustic measurements. Using the novel catheter, we show how optoacoustic measurements can determine variations of blood absorption, leading to accurate quantification of the detected NIRF signals at different hematocrit values.

show abstract

Automatic segmentation of intravascular ultrasound images: A texture-based approach

Cited by 28 publications

References 22 publications

Snakes based segmentation of the common carotid artery intima media

Snakes based segmentation of the common carotid artery intima media

Automatic quantification of histological studies in allergic asthma

Optoacoustic sensing of hematocrit to improve the accuracy of hybrid fluorescence‐ultrasound intravascular imaging

Contact Info

Product

Resources

About