Noise detection in 7T functional magnetic resonance imaging using artificial intelligence

• Grant number: 23/10867-3
• Support Opportunities: Scholarships in Brazil - Post-Doctoral
• Start date: September 01, 2023
• End date: February 29, 2024
• Field of knowledge: Health Sciences - Medicine - Medical Radiology
• Principal Investigator: Giovanni Guido Cerri
• Grantee: Mina Cintho Ozahata
• Host Institution: Faculdade de Medicina (FM). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated research grant: 21/11905-0 - Center of Science, Technology and Development for innovation in Medicine and Health: inLab.iNova, AP.CCD

Abstract

Functional Magnetic Resonance Imaging (fMRI) technique allows for the indirect estimation of brain connectivity by measuring the oscillation of the magnetic resonance (MR) signal caused by changes in blood oxygenation during brain activity. The assessment of connectivity among different brain regions relies on analyzing the correlation between temporal signal series in each region, which oscillates in accordance with brain activity. However, there are other factors that can contribute to signal fluctuation, such as physiological noise, electronic and instrumental noise stemming from the application of radiofrequency (RF) pulses and high-amplitude magnetic field gradients.In this grant proposal, we propose the development of a solution to assist researchers in reducing noise in resting-state fMRI signals. The aim is to enhance the differentiation between brain activation signal and noise, leading to increased sensitivity and specificity in studies of resting-state fMRI brain connectivity. This solution will be grounded in an Artificial Intelligence (AI) algorithm. We will gather both post-mortem and in vivo image series, encompassing various temporal and spatial resolutions, from brains of different sizes and age groups. This comprehensive image library, comprising structural and temporal data, will serve as the foundation for our approach. (AU)