This paper addresses the confidentiality, integrity, and availability (CIA) aspects of cloud-based medical image repositories. It proposes a two-tier security framework for medical images, including an additional security layer for the Region of Interest (RoI). The key generation process uses a 3-D Lorenz chaotic attractor to increase the keyspace, making statistical attacks infeasible. Deoxyribonucleic Acid (DNA) coding is employed for image diffusion, achieving high entropy and low correlation. Text encryption is used for metadata to ensure client authentication privacy. Encrypted metadata and 320 bits are embedded in the Non-Region of Interest (NRoI) part of the image using random pixel indexes generated by a 1D Tent map. A Graphical User Interface (GUI) is developed using Python 3.8 to support non-technical users and medical practitioners. The proposed security framework ensures a complete CIA triad for medical image repositories in the cloud, enhancing data security and integrity.This paper addresses the confidentiality, integrity, and availability (CIA) aspects of cloud-based medical image repositories. It proposes a two-tier security framework for medical images, including an additional security layer for the Region of Interest (RoI). The key generation process uses a 3-D Lorenz chaotic attractor to increase the keyspace, making statistical attacks infeasible. Deoxyribonucleic Acid (DNA) coding is employed for image diffusion, achieving high entropy and low correlation. Text encryption is used for metadata to ensure client authentication privacy. Encrypted metadata and 320 bits are embedded in the Non-Region of Interest (NRoI) part of the image using random pixel indexes generated by a 1D Tent map. A Graphical User Interface (GUI) is developed using Python 3.8 to support non-technical users and medical practitioners. The proposed security framework ensures a complete CIA triad for medical image repositories in the cloud, enhancing data security and integrity.