Cardiac Cycle Phase Estimation in 2-D Echocardiographic Images Using an Artificial Neural Network

Abstract

This paper proposes a new hybrid approach to estimate the cardiac cycle phases in 2-D echocardiographic images as a first step in cardiac volume estimation. We focused on analyzing the atrial systole and diastole events by using the geometrical position of the mitral valve and a set of three image features. The proposed algorithm is based on a tandem of image processing methods and artificial neural networks as a classifier to robustly extract anatomical information. An original set of image features is proposed and derived to recognize the cardiac phases. The aforementioned approach is performed in two denoising scenarios. In the first scenario, the images are corrupted with Gaussian noise, and in the second one with Rayleigh noise distribution. Our hybrid algorithm does not involve any manual tracing of the boundaries for segmentation process. The algorithm is implemented as computer-aided diagnosis (CADi) software. A dataset of 150 images that include both normal and infarct cardiac pathologies was used. We reported an accuracy of 90 % and a 2 ± 0.3 s in terms of execution time of CADi application in a cardiac cycle estimation task. The main contribution of this paper is to propose this hybrid method and a set of image features that can be helpful for automatic detection applications without any user intervention. The results of the employed methods are qualitatively and quantitatively compared in terms of efficiency for both scenarios.

Texture Spectrum Coupled with Entropy and Homogeneity Image Features for Myocardium Muscle Characterization

“…He and Wang [15,16] showed that any central pixel having an intensity value I0 is surrounded in a 3 × 3 neighborhood by eight pixels having the intensities {I1,I2,…,I8}. The TU is defined as an eight elements set {E1, E2,…,E8} given as: (1) where () Ii is the intensity of the pixel gray levels at the coordinates 1,...,8 i  inside the matrix of the Region of Interest (ROI). It should be noted that the element Ei is placed in the same position as the pixel i.…”

“…Distinctive echocardiographic features of myocardial pathologies related to the texture appearance have been discussed in several studies [1][2][3]. Bibicu and Moraru [1] proposed a hybrid approach to estimate the cardiac cycle phases in 2D echocardiographic images based on geometrical position of the mitral valve and on a set of three image features. The Artificial Neural Networks (ANN) method has been used as a classifier to select the requested anatomical information.…”

“…In texture analysis, the entropy and homogeneity are defined as the first-order textural features and the second-order textural features respectively, based on the relationship between pixel pairs [18][19][20]. In previous studies [1,21], textural image features have been used for segmentation or classification. Thus, the standard deviation image features computed in k × k (k = 3, 5, 7) masks are combined with the gray-level information to extract the shapes of breast and liver cysts from ultrasound images.…”

Máquinas de Aprendizado Extremo para Classificação Online de Eventos no detector ATLAS

“…Sistemas baseados em RNA vêm sendo utilizados com sucesso em diferentes aplicações como o suporteà decisão em inspeções não-destrutivas [9], diagnóstico biomédico [10] e seleção online de eventos em detector de partículas [5].…”

Novel approach for automatic mid‐diastole frame detection in 2D echocardiography sequences for performing planimetry of the mitral valve orifice

“…The authors of [6, 7] recognised the end‐systole and end‐diastole frames using the high‐dimensional data visualisation methods such as dimensionality reduction algorithms, which mapped the high dimensional of echocardiography image to low‐dimensional space. The authors of [8–10] carried out the further development of artificial neural networks and machine learning algorithms for determining cardiac phases automatically. The real‐time estimation of the cardiac cycle phase was the advantage of the introduced methods of these authors.…”