Quantum-Inspired Hyperchaotic Bio-DNA Image Encryption for Real-Time Medical Security
Harshit Sharma, Simran Jot Kaur
Abstract
The secure transmission of medical imagery in Internet of Medical Things (IoMT) environments is a critical requirement due to the high sensitivity and volume of diagnostic data shared over public networks. This paper proposes a novel, quantum inspired hyperchaotic image encryption framework that integrates adaptive Bio DNA mutation and multi domain transformation to protect medical data, particularly DICOM compliant CT and X ray images. The approach emulates quantum behavior via superposition mimicking bit plane diffusion and Grover like iterative scrambling, combined with entropy driven key initialization. A 3D Logistic Sine chaotic map is used to generate highly sensitive pseudo random sequences for spatial, frequency, and DNA domain encryption. An adaptive DNA mutation mechanism dynamically selects encoding rules based on local entropy and texture gradients, enhancing confusion and diffusion. The encryption pipeline applies Discrete Wavelet Transform (DWT) for frequency decomposition, followed by hyperchaotic scrambling and DNA domain permutation. Extensive experiments on multiple public and clinical datasets demonstrate strong resistance against differential, statistical, and brute force attacks, achieving average NPCR of 99.6451%, UACI of 33.4938%, and entropy scores approaching the ideal value of 8.0. The algorithm executes in 1.06 seconds on Raspberry Pi 4B for 512×512 images, confirming real time viability on embedded hardware. Ablation studies validate the contribution of adaptive DNA and multi domain architecture to security and performance. The proposed method achieves an average entropy of 7.9991, NPCR above 99.64%, and runtime of 1.06s on Raspberry Pi 4B, demonstrating its robustness, real time suitability, and scalability for IoMT based medical imaging.