Hiding Medical Information in Brain MR Images Without Affecting Accuracy of Classifying Pathological Brain.

Swagatika Devi,Manmath Narayan Sahoo,Khan Muhammad,Weiping Ding,Sambit Bakshi

DOI: https://doi.org/10.1016/j.future.2019.01.047

IF: 7.307

2019-01-01

Future Generation Computer Systems

Abstract:This work aims at validating the effect of steganography on a Computer-Aided Diagnosis (CAD) system for segregating pathological brain from normal brain from seeing MR images. Usually MR images of the brain achieves significant accuracy in classifying “normal” and “abnormal” brains. However, the brain images labeled with patient information is a severe threat to privacy preservation of the medical data, especially when the medical data is a part of Internet of Medical Things (IoMT) and travels over the network. Therefore, our article investigates on a two-fold objective i.e. ensuring visual hiding of steganographic data containing identity of the patient and at the same time retaining almost similar classification accuracy towards detecting a diseased MR brain image. In this approach, the identity of the patient is hidden in the image itself by employing a spatial steganographic algorithm on the T2-weighted MR images, thereby ensuring minimal chance of misuse of the private medical data. Two MR image datasets collected from Harvard whole brain Atlas have been used to validate our framework comparing against some well performing classification techniques. The classification techniques are applied on normal as well as stego images. It is concluded that using a steganographic algorithm on the MR images does not much affect the classification rate and yield equally satisfactory performance.