Adaptive image inversion of contrast 3D echocardiography for enabling automated analysis. (1st August 2015)

• Record Type: Journal Article
• Authors: Shaheen, Anjuman; Rajpoot, Kashif
• Is Part Of: Computers in biology and medicine. Volume 63(2015)
• Page Start: 99
• Page End: 107
• Publication Date: 2015-08-01
• Subjects: Contrast 3D echocardiography -- 3D echocardiography -- Image inversion -- LV segmentation -- Histogram analysis; Medicine -- Data processing -- Periodicals; Biology -- Data processing -- Periodicals; 610.285
• DOI: 10.1016/j.compbiomed.2015.05.009
• Languages: English
• ISSNs: 0010-4825

Abstract

Abstract: Introduction: Contrast 3D echocardiography (C3DE) is commonly used to enhance the visual quality of ultrasound images in comparison with non-contrast 3D echocardiography (3DE). Although the image quality in C3DE is perceived to be improved for visual analysis, however it actually deteriorates for the purpose of automatic or semi-automatic analysis due to higher speckle noise and intensity inhomogeneity. Therefore, the LV endocardial feature extraction and segmentation from the C3DE images remains a challenging problem. Methods: To address this challenge, this work proposes an adaptive pre-processing method to invert the appearance of C3DE image. The image inversion is based on an image intensity threshold value which is automatically estimated through image histogram analysis. Results: In the inverted appearance, the LV cavity appears dark while the myocardium appears bright thus making it similar in appearance to a 3DE image. Moreover, the resulting inverted image has high contrast and low noise appearance, yielding strong LV endocardium boundary and facilitating feature extraction for segmentation. Conclusions: Our results demonstrate that the inverse appearance of contrast image enables the subsequent LV segmentation. Highlights: Contrast 3d echocardiography suffers from high speckle. The automated analysis from contrast 3d echocardiography is a difficult task. Contrast image inversion is proposed to match appearance with non-contrast image. An automatic threshold estimation method is proposed to enable image inversion. The contrast image inversion enables LV segmentation using non-contrast methods.