A Cryptography-Based Framework for Securing Medical Imaging Data in Robotic Systems

Medical imaging data is crucial in robotics-assisted diagnostics and interventions, particularly in neurology. Ensuring the security and privacy of this sensitive data is increasingly challenging in the context of robotic systems. This study presents a robust framework to secure MRI brain images using the Advanced Encryption Standard (AES) algorithm, tailored for use in robotics applications. The AES algorithm is chosen for its high level of data confidentiality and integrity, making it well-suited for protecting sensitive medical information in robotic systems. Encrypted MRI images are decrypted and then processed by a convolutional neural network (CNN), specifically AlexNet, integrated into the robotic platform to classify the images as either tumorous or healthy. The classification performance is thoroughly evaluated using confusion matrices and performance plots, assessing the accuracy of the system. The project is implemented using Python, and the results demonstrate significant improvements over recent advancements in this field. This study not only enhances the security of medical imaging data within robotics applications but also contributes to improving the precision and reliability of robotic diagnostic systems.