A 3D approach to understanding heterogeneity in early developing autisms

This ShinyApp refers to the paper Mandelli, Severino et al., 2024 , which describes how autism could be divided into two subtypes: Type I and Type II , that emerge during Development . This first-level major distinction between two type of autims could be extremely useful in explaining developmental, biology, treatment- and outcome-relevant heterogeneity within the autism population.
In this work, to avoid any possible bias given by the adherence to a specific theory, Type I and Type II subtypes are identified in a data driven fashion using cluster analysis in the context of stability-based relative validation, please see Landi et al., 2021

In this web app you will be able classify an individual into the Type I or Type II, by inserting Mullen and Vineland results of his/her assessment taken between 24 and 68 months of age
The side tabs on this app will allow you to insert data from a single individual (e.g., Single Subject tab), or to input a text file to process a batch of individuals (e.g., Database tab)

AUTISMS-3D theoretical model - A3D

The A3D theoretical model proposes that autism can be divided into two subtypes, distinguished by the level of skills in non-core features of autism, such as: Language, Intellectual, Motor, and Adaptive skills (LIMA). The A3D model suggests that there is one subtype of autism with significant disabilities in the LIMA features (Type I) and another subtype without these difficulties (Type II) . These subtypes are supposed to have different longitudinal outcomes, as their developmental trajectories are divergent (see Figure 1 below). Additionally, the A3D model theorizes that the two subtypes can be distinguished at a neurobiological level (see Figure 2 below).
Future research may apply this subtyping to conduct further studies and test the model’s validity in indicating significant differences in autism. For a comprehensive understanding of the theoretical model, please refers to our paper .