Segmentation of mitral valve leaflets in ecocardiographic images

• Grant number: 20/11319-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: November 01, 2020
• End date: October 31, 2021
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal Investigator: Regina Célia Coelho
• Grantee: Gabriele Fernandes Garcia
• Host Institution: Instituto de Ciência e Tecnologia (ICT). Universidade Federal de São Paulo (UNIFESP). Campus São José dos Campos. São José dos Campos , SP, Brazil

Abstract

The mitral valve is an essential structure for cardiac function because it is responsible for the passage of blood from the left atrium to the left ventricle through its two leaflets. The most common diagnosis regarding the functioning of this valve is made from echocardiographic exams, considering its ability to capture the valve's rapid movement, its affordable price, and non-ionizing characteristics for the patient. The processing of echocardiogram images has a fundamental function in diagnosing diseases related to the mitral valve, introducing greater precision of information for interpretation and medical decision making, especially through the segmentation technique. Due to the relevance of segmentation of the mitral valve to aid medical diagnosis, the purpose of this project is to segment the fully open and fully closed leaflets of the valve in the long axis parasternal cut of echocardiogram images. Kohonen's self-organizing maps and Image Processing techniques will be used for pre- and post-processing images from three databases. Finally, the results will be validated using the Dice and Jaccard coefficient to compare the segmentation results with the contours marked by a specialist.