No abstract

No abstract

No abstract

Image segmentation is an important technology for computer-aided diagnosis. Purpose of medical image segmentation is to separate the images' regions with special meanings and to make the segmentation as close as possible to the anatomical structure so as to provide a reliable foundation for clinical diagnosis and treatment and also for pathological study. Due to the complexity of human anatomy, the irregularity of organ and tissue shapes as well as the diversity between individuals, general medical image segmentation methods cannot be directly applied to achieve the ideal segmentation effect. Besides, medical images have different features which cannot endow segmentation methods with universality. Thus, a more targeted medical image segmentation method is needed. This paper is targeted at studying an effective method for segmenting cardiac images, namely, the segmentation of coronary arteries and the segmentation of ventricle and atrium (VA). According to the characteristics and special needs of medical images, a new method for cardiac image segmentation was proposed. This method combines contrast enhancement, threshold segmentation, morphological algorithm and level set-based active contour model. Furthermore, this method improves its own problems to ultimately achieve the algorithm in MATLAB platforms, obtaining a better segmentation result.

Abstract. Despite the application of contrast enhancement to image enhancement in medical image processing and in spite of a good enhancement effect achieved, since the priori knowledge of the relevant application domain is left out, the medical images affected by noise have yet to be further enhanced. Priori knowledge-based contrast enhancement can well eliminate noise effect by extracting known part of mechanism from an object as priori knowledge and combining it with sample data to build a reliable sample model. This paper proposed a priori knowledge-based cardiac CT image block adaptive contrast enhancement algorithm, which can be used to process various characteristics of a cardiac CT image flexibly to general an ideal enhanced image.