Tracking large Anterior Mitral Leaflet displacements by incorporating optical flow in an active contours framework.

Echocardiography is an important tool to detect early evidence of mitral valve degradation associated with rheumatic heart disease. The segmentation and tracking of the Anterior Mitral Leaflet helps to quantify the morphologic valve anomalies, such as the leaflet thickening, shape and the mobility changes. The tracking of this leaflet throughout the cardiac cycle is still an open challenge in the research community. The widely used active contours segmentation framework fails when faced with large leaflet displacement. In this work, we propose the integration of optical flow in an open-ended active contour framework to address this difficulty. This additional information promotes solutions with contours next to high leaflet displacements, resulting in superior performance. The algorithm was tested on 9 fully annotated real clinical videos, acquired from the parasternal long axis view. The algorithm is compared with our previous work. Results show a clear improvement in situations where the leaflet exhibits large displacement or irregular shapes, with an average error of 4.5 pixels and a standard deviation of 2 pixels.

Tracking of the anterior mitral leaflet in echocardiographic sequences using active contours.

Echocardiography assessment of cardiac valves plays a vital role in the diagnosis of rheumatic heart disease. In the vast majority of cases, the mitral valve gets affected, leading to the thickening of its leaflets that may result in the fusion of their tips. This changes the appearance and reduces the mobility of the leaflets, which also reduce the heart efficiency. Quantifying such parameters provides diagnostic insight. To achieve that, the first step is to identify and then track fast moving leaflets. This work is focused on Anterior Mitral Leaflet (AML) tracking. Open ended active contours are employed in this work by removing its boundary conditions. The external and internal energy of the contour is modified that extend the capture range, improve snake energy and encourages the leftmost end point of the contour to converge on the moving tip of the AML. Results show that contour points are tracked accurately with an average error of 4.9 pixels and a standard deviation of 2.1 pixels in 9 fully annotated normal sequences of real children clinical assessments.

How well can the fusion of Gabor filters and local binary patterns help in identifying gastric lesions?

Gastroenterology imaging is a diagnostic procedure that incorporates various computer vision challenges for the design of assisted diagnostic systems. The most typical challenge is the design of more adequate visual descriptors that can assist the classification algorithms in getting good diagnostic results. Literature shows that most of the texture descriptors for feature extraction from gastric lesions are based on Gabor filters or local binary patterns (LBP). Although good results are obtained, these techniques have their shortcomings. In this paper, we aim to explore the use of fusion of Gabor filters and LBPs for characterizing gastric lesions. The images are first subjected to Gabor filtering using isotropic Gabor filters, followed by extracting LBPs from the filtered images. We validate the performance of the descriptor on a novel gastroenterology dataset: the Post-MAPS dataset. Our results show that the proposed feature set outperforms the other methods that have been considered in this paper.

Segmentation of bones & MCP joint region of the hand from ultrasound images.

Rheumatic arthritis (RA) is an autoimmune disease that causes irreversible damage to joints and other physiological structures. The Metacarpophalangeal (MCP) joint is one of the first regions to suffer alterations. These alterations are visible with high frequency ultrasound devices, which are used to quantify inflammatory activity in the MCP due to RA. The accurate segmentation of the bone surface and the identification of the MCP capsule region remains a challenge in ultrasound image processing. In this article we aim to make a contribution to this problem by incorporating prior knowledge of the bone and joint regions anatomy into our segmentation algorithm. The log Gabor filter is used for speckle noise reduction and to extract ridge-like structures from the images, while the phase is left unchanged. After thresholding, scores are generated, based on the intensities and areas of the resulting regions, enabling the selection of the structure that best matches the bone. Finally, segmented joint bones are processed to calculate the initial seeds of joint capsule region. Experimental results demonstrate the accuracy of the proposed segmentation algorithm. The mean pixel error between the automatic segmentation and the reference images were 4.4 pixel. The bone regions not segmented were, on average, 5.4%.

A multi-spot exploration of the topological structures of the reconstructed phase-space for the detection of cardiac murmurs.

Acoustic heart signals are generated by a turbulence effect created when the heart valves snap shut, and therefore carrying significant information of the underlying functionality of the cardiovascular system. In this paper, we present a method for heart murmur classification divided into three major steps: a) features are extracted from the heart sound; b) features are selected using a Backward Feature Selection algorithm; c) signals are classified using a K-nearest neighbor's classifier. A new set of fractal features are proposed, which are based on the distinct signatures of complexity and self-similarity registered on the normal and pathogenic cases. The experimental results show that fractal features are the most capable of describing the non-linear structure and the underlying dynamics of heart sounds among the all feature families tested. The classification results achieved for the mitral auscultation spot (88% of accuracy) are in agreement with the current state of the art methods for heart murmur classification.

Detecting melanoma in dermoscopy images using scale adaptive local binary patterns.

Recent advances in the area of computer vision has led to the development of various assisted diagnostics systems for the detection of melanoma in the patients. Texture and color are considered as two fundamental visual characteristics which are vital for the detection of melanoma. This paper proposes the use of a combination of texture and color features for the classification of dermoscopy images. The texture features consist of a variation of local binary pattern (LBP) in which the strength of the LBPs is used to extract scale adaptive patterns at each pixel, followed by the construction of a histogram. For color feature extraction, we used standard HSV histograms. The extracted features are concatenated to form a feature vector for an image, followed by classification using support vector machines. Experiments show that the proposed feature set exhibits good classification performance comparing favorably to other state-of-the-art alternatives.