3D Medical Image Segmentation
Marwa Ibrahim Shams;
Abstract
Recently, automated analysis of medical images becomes important for easier and faster clinical diagnosis. Identifying human organs is the key component for such analysis, i.e., segmentation of the anatomical structures from medical images.
Coronary arteries segmentation in three-dimensional (3D) images has gained wide interest in old and recent scientific research and is regarded as a fundamental step in evaluating the degree of Coronary Artery Disease (CAD) in cardiac clinical diagnosis and surgical planning. Thus, various methods have been developed for segmenting coronaries from different cardiac imaging modalities.
Previously developed segmentation methods were designed in a way that can address the challenging task of coronary arteries. The challenges of coronary segmentation can be summarised in four points: the small size of coronaries, structures attached to coronaries has similar intensity and intensity, shape variations along the vessels and presence of calcifications.
The research problem of coronary segmentation was divided into three parts. First, enhancing the 3D input images using vessel enhancement techniques which make it easier to detect and extract coronary vessel regions in next steps. Second, recognizing and segmenting coronaries using a proper segmentation method that can handle the changes in intensity and geometry along coronaries. Third, using the resulted coronary vessel tree for detecting and quantifying stenoses (narrowness).
A framework for a coronary segmentation, stenoses detection and quantification system is proposed along with a comprehensive overview of the state-of-art coronary segmentation algorithms.
The proposed coronary segmentation framework was divided into three main parts: enhancement and preprocessing, coronary segmentation, and stenoses detection and quantification. In enhancement, input CTA images are enhanced by removing calcifications using thresholding, making coronary regions more
Coronary arteries segmentation in three-dimensional (3D) images has gained wide interest in old and recent scientific research and is regarded as a fundamental step in evaluating the degree of Coronary Artery Disease (CAD) in cardiac clinical diagnosis and surgical planning. Thus, various methods have been developed for segmenting coronaries from different cardiac imaging modalities.
Previously developed segmentation methods were designed in a way that can address the challenging task of coronary arteries. The challenges of coronary segmentation can be summarised in four points: the small size of coronaries, structures attached to coronaries has similar intensity and intensity, shape variations along the vessels and presence of calcifications.
The research problem of coronary segmentation was divided into three parts. First, enhancing the 3D input images using vessel enhancement techniques which make it easier to detect and extract coronary vessel regions in next steps. Second, recognizing and segmenting coronaries using a proper segmentation method that can handle the changes in intensity and geometry along coronaries. Third, using the resulted coronary vessel tree for detecting and quantifying stenoses (narrowness).
A framework for a coronary segmentation, stenoses detection and quantification system is proposed along with a comprehensive overview of the state-of-art coronary segmentation algorithms.
The proposed coronary segmentation framework was divided into three main parts: enhancement and preprocessing, coronary segmentation, and stenoses detection and quantification. In enhancement, input CTA images are enhanced by removing calcifications using thresholding, making coronary regions more
Other data
| Title | 3D Medical Image Segmentation | Other Titles | تجزئة الصور الطبية ثلاثية الأبعاد | Authors | Marwa Ibrahim Shams | Issue Date | 2017 |
Recommend this item
Similar Items from Core Recommender Database
Items in Ain Shams Scholar are protected by copyright, with all rights reserved, unless otherwise indicated.