Interdisciplinary research of machine learning and artificial techniques for development of signal, image and natural language processing technologies

• Grant number: 22/03243-0
• Support Opportunities: Research Grants - Initial Project
• Duration: February 01, 2023 - January 31, 2028
• Field of knowledge: Engineering - Electrical Engineering - Telecommunications
• Principal Investigator: André Kazuo Takahata
• Grantee: André Kazuo Takahata
• Host Institution: Centro de Engenharia, Modelagem e Ciências Sociais Aplicadas (CECS). Universidade Federal do ABC (UFABC). Ministério da Educação (Brasil). Santo André , SP, Brazil
• Associated researchers: Celso Setsuo Kurashima ; Diogo Coutinho Soriano ; Fabio Luiz Franceschi Godinho ; Fernando Silva de Moura ; Lilian Berton ; Margarethe Born Steinberger-Elias ; Maria Sheila Guimarães Rocha ; Priscyla Waleska Targino de Azevedo Simões ; Ricardo Suyama
• Associated scholarship(s): 23/17178-9 - Voicebot for hybrid care based patient telemonitoring platforms, BP.MS; 23/13729-0 - Electrical Impedance Tomography for Stroke Diagnosis Using Deep Neural Networks, BP.DR

Abstract

This project aims to consolidate an interdisciplinary research group with researchers in the fields of information engineering, biomedical engineering, linguistics, computer science and medicine. The objective of this group is to develop machine learning and artificial intelligence engineering applications for signal, image and natural language processing, prioritizing health applications. This project is based in work that is already being carried out by the proponent such as applying machine learning for defining Parkinson's disease biomarkers from local field potential (LFP) signals acquired during implantation of deep brain stimulation (DBS) devices and use of deep neural networks to detect cerebrovascular accident (CVA) in electrical impedance tomography. Also, in natural language processing research, we propose to study the characterization of texts on COVID-19, in order to detect biomedical words with deep neural netowrks, and to develop voicebots for hybrid care platforms based on patient telemonitoring. (AU)