The secure transmission of sensitive medical data, such as structural MRI (sMRI) images, is a vitalrequirement for protecting patient confidentiality and maintaining data integrity in modern healthcareand research applications. This study introduces a comprehensive framework that employsFernet key encryption as a core component to ensure the secure encoding and transmission of Smridata, specifically in the context of autism spectrum disorder (ASD) prediction. Fernet encryption, asymmetric encryption technique, provides end-to-end confidentiality by encrypting the data with asecret key, ensuring that only authorized parties with the correct decryption key can access the originaldata. Within the framework, sMRI data is securely encrypted before transmission to preventunauthorized access or tampering during the transfer process. Upon reaching the intended recipient,the encrypted data is decrypted, restoring its original form for further analysis. Subsequently,vision transformers are applied to the reconstructed sMRI data to evaluate the likelihood of ASDin individuals, offering valuable diagnostic insights while maintaining rigorous security protocols.This approach ensures the confidentiality, integrity, and usability of sensitive medical data, makingit a robust and practical solution for secure data transmission. By leveraging the simplicity andreliability of Fernet encryption, this framework provides a scalable and efficient method to safe-guard medical data while enabling its utilization for advanced diagnostic purposes in healthcare andresearch settings.