Improving the Spatial Resolution of the TIE Reconstructed Image with Neural Networks

• Grant number: 22/06874-1
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: January 01, 2023
• End date: December 31, 2023
• Field of knowledge: Engineering - Biomedical Engineering - Bioengineering
• Principal Investigator: Marcos de Sales Guerra Tsuzuki
• Grantee: Alexandre Silva Bastos de Almeida
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

From information on electrical potential at the perimeter of the cross section of an object (gathered by the placement of electrodes), an Electrical Impedance Tomography (EIT) system must be able to estimate a distribution of conductivities within this section, in a non invasive way. The reconstructed image has admittedly low spatial resolution. It is intended to improve the spatial resolution with the use of deep neural networks based on super resolution. Two networks will be studied: 1. based on residuals; and, 2. U-Net based. This will allow images to be reconstructed with higher resolution in reduced processing time.