A dual layer security scheme for medical images using Hessenberg and singular value decompositions

Due to the recent advancement in the field of the Internet of Medical things (IoMT). To facilitate doctors and patients, in the process of diagnosis and treatment, the medical imaging equipment is connected to the IoMT. During communication over the network, these medical images are subjected to various threads. In this work, we have proposed a dual-layer data confidentiality scheme, firstly it encrypts the secret medical images followed by a data hiding scheme. The encryption scheme possesses diffusion and confusion, for confusion the encryption scheme utilizes logistic and tent maps for the generation of S-boxes. For data hiding, it utilizes Hessenberg and singular value decomposition (SVD). The proposed scheme is applied to highly correlated medical images. The proposed technique provides dual security to the confidential information and makes it difficult for the intruder to extract the confidential information. The encryption scheme is evaluated by using the standard performance indicators including statistical analysis, differential analysis, and NIST analysis, etc. The encrypted images have the highest practically achievable entropy of 7.999 which is closest to the ideal value of 8. The data hiding scheme is evaluated by using statistical analysis, Distance-based analysis, analysis based on pixel difference, and information theory. Both the analysis of encryption and data hiding are satisfactory and the results show the strength of the dual-layer security scheme.