A Novel Approach for Recognition of Autism Spectrum Disorder based on GraphSAGE

Diagnosing Autism Spectrum Disorder can be quite challenging due to its nature and diverse presentation styles; hence the early detection is key for effective intervention strategies to be implemented in atimely manner. There has been a lot of talk lately aboututilizing deep learning algorithms to enhance the finding of ASD by analyzing neuroimaging data more effectively. To overcome the limitations of prevailing methods in this area of research this study introduces an innovative method known as the Multi view United Transformer Block-GraphSage or MVUT_GraphSage, for short. MVUT_GraphSage makes use of the advantages provided by view learning and aggregation processes to uncover delicate shapes hidden within the MRI data which is structural and functional. After conducting an examination using the ABIDE dataset it was found that MVUT_GraphSage performs better than the current Multi View Site with the Graph Convolution Network (MVS_GCN) resulting in a +2.57% increase in accuracy rate which showcases the efficiency of the proposed model when it comes to recognizing ASD cases accurately It is believed that this better accuracy and consistency in identifying ASD through MVUT_GraphSage could lead to earlier intervention and better support for individuals, with ASD Moreover MVUT_GraphSage ability to be understood makes it easier for deep learning models to connect with clinical insights by assisting in identifying biomarkers linked to ASD. In the end. All things considered was ultimately successful in enhancing our knowledge about ASD and how to practically handle it which could lead to enhancing the results and eminence of life, for persons impacted by it.