HPDH-MI: A High Payload Data Hiding Technique for Medical Images Based on AMBTC

In the realm of electronic health (eHealth) services powered by the Internet of Things (IoT), vast quantities of medical images and visualized electronic health records collected by IoT devices must be transmitted daily. Given the sensitive nature of medical information, ensuring the security of transmitted health data is paramount. To address this critical concern, this paper introduces a novel data hiding algorithm tailored for Absolute Moment Block Truncation Coding (AMBTC) in medical images, named HPDH-MI (High Payload Data Hiding for Medical Images). The proposed method embeds secret data into the AMBTC compression code inconspicuously to avoid detection by malicious users. It achieves this by first classifying AMBTC compressed blocks into four categories—flat, smooth, complex I, and complex II—using three predetermined thresholds. A 1-bit indicator, based on the proposed grouping strategy, facilitates efficient and effective block classification. A data embedding strategy is applied to each block type, focusing on block texture and taking into account the symmetric features of the pixels within the block. This approach achieves a balance between data hiding capacity, image quality, and embedding efficiency. Experimental evaluations highlight the superior performance of HPDH-MI. When tested on medical images from the Osirix database, the method achieves an average image quality of 31.22 dB, a payload capacity of 225,911 bits, and an embedding efficiency of 41.78%. These results demonstrate that the HPDH-MI method not only significantly increases the payload for concealing secret data in AMBTC compressed medical images but also maintains high image quality and embedding efficiency. This makes it a promising solution for secure data transmission in telemedicine, addressing the challenges of limited bandwidth while enhancing steganographic capabilities in eHealth applications.