Organoid intelligence: a promising paradigm for autism spectrum disorder research

Autism spectrum disorder (ASD) is a multifactorial neurodevelopmental condition marked by heterogeneity in its clinical presentations, which complicates both its diagnosis and treatment. Despite advancements, the molecular mechanisms underlying ASD remain largely unclear. In recent years, two independent yet highly promising technological domains have made significant advancements: organoids and artificial intelligence (AI). Brain organoids can simulate critical processes of human neurodevelopment in both physiological and pathological states, providing an unprecedented window into the mechanisms and potential therapeutics of ASD. Simultaneously, AI has demonstrated formidable capabilities for processing and analyzing large-scale, high-dimensional biomedical data, and has been effectively applied to ASD imaging analysis, genomics research, and behavioral data interpretation. However, these two fields have developed mainly in parallel, leaving their potential for cross-disciplinary integration untapped in ASD research. This review aims to fill this gap by reviewing the current status and limitations of both fields in the context of ASD, and to elucidate a promising research paradigm that integrates AI with brain organoids, called "organoid intelligence", to enhance our understanding of predictive models, unravel ASD pathogenesis, and develop more effective and individualized therapies.