Age of machine learning: new trends in autism spectrum disorder prediction

In recent years, there has been an increase in the incidence of autism spectrum disorder (ASD), its pathogenesis remains unknown, and there are no effective treatments available. Early identification of individuals at risk enables early targeted intervention, which improves outcomes. Through the integration of artificial intelligence and the medical field, researchers can establish a machine learning (ML) risk prediction model to estimate the risk of ASD. Currently, a variety of risk models have been developed using multiple factors, such as genetic background, gaze behavior, adverse conditions during pregnancy and childbirth, magnetic resonance imaging of the brain, and intestinal microbial composition, to predict ASD. These ML prediction models have shown some reliability in predicting ASD risk. In the future, ML prediction models for ASD will present significant challenges and opportunities, potentially helping identify drug targets for developing novel therapies to alleviate ASD symptoms and enable precision medicine.