S-TPE: Usability-Enhanced Thumbnail-Preserving Cryptosystem Based on S-Boxes for Securing IoT-Based Medical Images

With the growth of the Internet of Things (IoT), medical and multimedia data increasingly rely on visuals. Thus, in the Internet of Medical Things (IoMT), encryption safeguards sensitive images but also introduces noise, which makes them harder to recognize. Consequently, this decreases availability and hinders rapid image identification. Many researchers use Thumbnail Preserving Encryption (TPE) for visibility needs, but most schemes suffer from less efficiency, high data expansion, and low security. This work addresses these by proposing a lightweight cryptographic method that builds strong S-boxes using the 2D Henon map and Grey Wolf Optimization (GWO). These optimized S-boxes present robust cryptographic features like balanceness and nonlinearity, which are difficult to achieve. The first S-box has high nonlinearity value 110. The proposed TPE scheme uses bitwise swapping and substitution-permutation with the Rank(.) and Rank-1(.) functions utilizing optimized S-boxes to secure the image while preserving the thumbnail. This method meets the need for fast processing, ideal thumbnails, full decryption, and strong protection where current TPE methods fail. It ensures that the encrypted image is difficult to understand and resists various attacks, including statistical, chosen-plaintext, and differential attacks. Simulation reveals good image quality, speed, and security, making it fit for secure IoMT healthcare use.