CDLAKS: Chaotic DNA-Based Lightweight Authenticated Key-Dependent Steganographic Framework for Secure Transmission of Medical Images in Iot Environments

With the rapid deployment of Internet of Things (IoT) technologies in healthcare, the secure transmission of medical images and associated patient data has become a critical challenge due to strict resource constraints and increasing privacy threats. This paper presents a chaotic DNA-based lightweight authenticated key-dependent steganographic cryptosystem, that is known as CDLAKS system, for the secure and authenticated transmission of DICOM medical images in IoT environments. The proposed system incorporates multi-map chaotic key generation, DNA-based adaptive encoding, Arnold Cat map scrambling, and key-sensitive Least Significant Bit (LSB) steganography under a single 256-bit secret key. Data integrity and authentication are ensured by embedding a SHA-1 hash of the DNA-encoded payload on a cover image, which can be verified at the receiver side. Extensive investigation over standard medical images, such as CT, MRI, and X-ray modalities, reveals high imperceptibility, with Peak Signal-to-Noise Ratio (PSNR) values of up to 79.8 dB and Structural Similarity Index (SSIM) values close to unity. Differential attack analysis using the Number of Pixel Change Rate (NPCR) and Unified Average Changing Intensity (UACI) confirms that the proposed scheme controls pixel modifications, making it well-suited for steganographic applications. The statistical test results obtained from the NIST Statistical Test Suite highlight the randomness of the 256-bit secret key. Furthermore, a hardware feasibility evaluation on the Spartan-6 FPGA has shown that the proposed design uses fewer than 12% of LUTs and 6% of memory and DSP blocks, thereby confirming its effectiveness for real-time and lightweight IoT healthcare systems. In summary, the CDLAKS scheme achieves an effective compromise among security, imperceptibility, authentication, and efficiency for secure medical image communication.