Structuring and deduplicating a database for supporting computer-aided diagnosis of Autism Spectrum Disorder

• Grant number: 24/21470-0
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: March 01, 2025
• End date: February 28, 2026
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computer Systems
• Principal Investigator: Fátima de Lourdes dos Santos Nunes Marques
• Grantee: Sarah Klock Mauricio
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil

Abstract

Computer-aided diagnosis schemes of Autism Spectrum Disorder (ASD) aim to classify individuals with typical development or ASD categories based on varied input data. In particular, an approach under development at the Laboratory of Computer Applications for Health Care ( Laboratório de Aplicações de Informática em Saúde - LApIS) of the School of Arts, Sciences, and Humanities at the University of São Paulo (USP), in collaboration with the Autism Spectrum Disorders Program (PROTEA) of the Institute of Psychiatry at the Hospital das Clínicas (IPq-HC) of the Faculty of Medicine of the University of São Paulo (FMUSP), utilizes eye-tracking data and anthropometric analysis of facial images to classify individuals as belonging or not belonging to the ASD spectrum. These algorithms provide a more accessible alternative applicable at earlier ages compared to traditional diagnostic methods such as the Autism Diagnostic Interview-Revised (ADI-R) and the Autism Diagnostic Observation Schedule (ADOS). This project aims to create, implement, and validate a database to support the development and use of these algorithms, as well as its integration with existing databases and the development of an interface to facilitate access to the data. Data is collected from various locations, increasing occurrence of duplication of registered individuals with similar but not identical data. Therefore, in addition to the database, this project aims to define and implement an approach for deduplication using demographic and facial biometric data to minimize inconsistencies caused by redundancies. The main expected results include the development of the database, an interaction interface, an adequate deduplication method, and the production of academic articles.