Segmentação de imagens médicas para criação de atlas anatômico aplicado à tomografia por impedância elétrica

• Grant number: 21/05566-9
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2021
• End date: August 31, 2022
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal Investigator: Marcos de Sales Guerra Tsuzuki
• Grantee: Jungeui Choi
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

In this research, 3D computed tomography (TC) images of the human chest will be automatically segmented. Bones, heart, lungs, airways, and skin will be segmented. These results will be used in other scientific initiation, where the segmented organs will be used in the creation of 3D meshes with electrodes. This 3D meshes will be used to create a dataset for training neural networks that will solve the problem of electrical impedance tomography. The images are in DICOM images that will be read and segmented by an algorithm based on the Image-Forest Transform Oriented for 3D images. This algorithm was developed by Prof. Dr. Paulo André Vechiatto de Miranda from the Institute of Mathematics and Statistics/USP. This algorithm requires the provision of seed regions. In this scientific initiation, algorithms for seed generation for the segmentation algorithm will be developed.(AU)