Toward an Automatic Left Atrium Localization Based on Shape Descriptors and Prior Knowledge
| Autor: | Saïd Mahmoudi, Amine Abbou, Mohammed Amine Chikh, Mohammed Ammar |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
medicine.medical_specialty
business.industry Left atrium Atrial fibrillation Anatomy medicine.disease Pump blood Automatic localization Mri image medicine.anatomical_structure Ventricle medicine.artery Internal medicine Pulmonary artery cardiovascular system Cardiology medicine Segmentation business |
| Zdroj: | Statistical Atlases and Computational Models of the Heart. Imaging and Modelling Challenges ISBN: 9783642542671 STACOM |
| DOI: | 10.1007/978-3-642-54268-8_5 |
| Popis: | The left atrium is one of the four chambers of the heart. It receives oxygenated blood from the lungs and pumps it into the left ventricle. This blood is then circulated to the rest of the body. In a healthy adult the left atrium pumps blood into the ventricle in a regular rhythm. In atrial fibrillation AF, the left atrium quivers in an abnormal rhythm and is no longer able to pump blood into the left ventricle efficiently. On the other hand MRI and CT are commonly used for imaging this structure. Segmentation can be used to generate anatomical models that can be employed in guided treatment and also more recently for cardiac biophysical modelling. For this reason, segmentation of the left atrium is a task with important diagnostic power. In this paper, we propose an automatic localization method in order to detect the left atrium in MRI images. Our method is based on shape descriptor and prior knowledge. For this purpose some descriptors are selected: circularity, area, the center of mass of each region, elongation factor, type factor. We propose also to use some prior knowledge as pulmonary artery position, and the left atrium position. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|