Scattering and Difraction through Oriented Systems: Pattern Recognition and Automated Modeling by Machine Learning and Artificial Intelligence

• Grant number: 25/03179-9
• Support Opportunities: Scholarships in Brazil - Doctorate (Direct)
• Start date: July 01, 2025
• End date: February 28, 2030
• Field of knowledge: Physical Sciences and Mathematics - Physics - Condensed Matter Physics
• Principal Investigator: Cristiano Luis Pinto de Oliveira
• Grantee: Gabriel Brito Granado
• Host Institution: Instituto de Física (IF). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

In this project, we propose the development of computational tools for analyzingand modeling Small-Angle Scattering (SAS) data, with emphasis on systems with ananisotropic intensity profile. A methodology is proposed that encompasses the applicationof artificial intelligence (AI) with machine learning (ML) for the automation of analysisprocedures, including pattern recognition in two-dimensional (2-D) images, sectorintegration and structural modeling of experimental data. With the increasing availabilityof high-resolution 2-D images at 4th generation synchrotrons, such as Sirius at theNational Laboratory of Synchrotron Light (LNLS), this project arises from the necessityfor tools that can deal with the large volume of data generated in these large centers. Themethodologies include the proposition of deep learning (DL) based neural networks(NNs) models for classifying one-dimensional (1-D) intensity profiles into suitablestructural models. Convolutional neural networks (CNNs) are proposed to improve theidentification of regions of interest in anisotropic 2-D profiles. Validation and testing ofevery model will be carried out using experimental data from oriented (and thereforeanisotropic) systems, such as oriented hair fibers and liquid crystals, in conjunction withsimulations. The main goal of the project is to provide innovative and efficient solutionsthat can automate the analysis of complex systems and promote advances in the field ofSmall-Angle Scattering. (AU)