SYSTEMATIC MAPPING: SCREENING SCALES FOR AUTISM SPECTRUM DISORDER AND MATHEMATICAL MODELING

Abstract

Early diagnosis of Autism Spectrum Disorder (ASD) is essential to enable timely therapeutic and educational interventions. However, the screening process still faces challenges related to the subjectivity of assessments, the diversity of available instruments, and uncertainties in the interpretation of results. This article presents a systematic mapping of the literature on instruments and scales used for ASD screening, their main limitations, and the use of technological and mathematical modeling approaches to support the diagnostic process. Publications from 2014 to 2025 were analyzed in the ACM Digital Library, IEEE Xplore, PubMed, Virtual Health Library, and CAPES Portal databases. The initial search identified 1,155 studies, of which 36 were included in the final mapping. The results revealed 26 instruments and limitations related to subjectivity, cross-cultural differences, costs, specialized training, and borderline cases. Approaches based on artificial intelligence, fuzzy logic, and expert systems were also identified. The findings highlight gaps in the representation of uncertainty and indicate the potential of mathematical modeling to complement existing instruments and support the interpretation of ASD screening results.