Application of machine learning techniques to determine the position of electrodes in electrical impedance tomography

• Grant number: 21/05565-2
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2021
• End date: June 30, 2022
• Field of knowledge: Engineering - Biomedical Engineering - Medical Engineering
• Principal Investigator: Marcos de Sales Guerra Tsuzuki
• Grantee: Eduardo Jubran Pascual
• Host Institution: Escola Politécnica (EP). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

From electrical potentials measured at electrodes in the perimeter of the cross section of an object, a system of Electrical Impedance Tomography (EIT) should be able to estimate an impedance distribution within this section, in a non-invasive way. In this research project, the possibility of also determining whether an electrode is positioned incorrectly and which electrode is it will be investigated. The possibility of identifying the variation in incorrect positioning will also be investigated. Classifiers will be used to identify the existence of an incorrectly positioned electrode and, if it exists, to identify the incorrectly positioned electrode. Several classification techniques will be tested, such as: random forest, decision trees and SVM. Techniques based on autoencoder and Siamese networks will also be tested. Autoencoders allow you to determine an encoding for the domain being explored. The Siamese networks favor the grouping between similar elements and the separation of the different elements. (AU)