ABSTRACT

The research created a novel reversible data hiding (RDH) strategy for disease diagnosis depending on Code Division Multiplexing (CDM) as well as machine learning techniques. The hidden information was embedded further into mid-frequency subcarriers of the ct images using CDM plus machine learning techniques after the patient records picture was translated into a spectral domain using the decimal digits decomposition method. The hidden information was implanted continuously thanks to the orthogonality of multiple extending phases used in the CDM method, and most of the components of extending processes were directly negated, resulting in a large information hiding potential with minimum picture degradation. Concurrently, the hidden information to be embedded was indicated by distinct disseminating patterns, and then only the recipient with an identical disseminating pattern as the transmitter could fully recover the secret information and reference picture, thereby improving the RDH's safety. Keywords: Reversible data hiding; Code Division Multiplexing; Machine learning
Publication date: 01/11/2021

https://ijbpas.com/pdf/2021/November/MS_IJBPAS_2021_NOV_SPCL1045.pdf

Download PDF
https://doi.org/10.31032/IJBPAS/2021/10.11.1045