Page 80 - 2024F

UEC Int’l Mini-Conference No.53 73

Data Hiding Components for Solving Information Security Issues in

DICOM Medical Images

Eng. Juan Eduardo Mosco Garcia, Ph.D. Hayaru Shouno
Department of Engineering
The University of Electro-Communications
Tokyo, Japan

Keywords: Data Hiding, Halftoning, DICOM, Medical Imaging, Information Security.

Abstract
Ensuring patient data protection in medical imaging is crucial for maintaining privacy and se-
curity. This research investigates various halftoning techniques and identifies Jarvis halftoning as
the most effective for image reconstruction, enhancing data hiding methods. A high-capacity data
hiding approach was applied to binary images by embedding binary characters into the modified
contours of Jarvis halftone images, increasing the amount of hidden information. To further im-
prove embedding capacity, artificial contours were introduced in specific pixel locations, optimizing
the halftoning process. The processed images were embedded into DICOM medical images using
a pseudo-random traversal algorithm, generating two encryption keys embedded in layers 12 and
16 of the image. The modified DICOM image was evaluated using the Structural Similarity Index
(SSIM). The extracted data allowed for the recovery of the modified Jarvis halftone image, which
was then converted back to grayscale to facilitate patient authentication. The results demonstrate
the technique’s ability to preserve image quality after data embedding and extraction. This study
highlights the potential of integrating advanced halftoning and data hiding strategies in medical
imaging, improving data security in the storage and transmission of sensitive patient information
while ensuring data integrity and confidentiality.

∗
The author is supported by JASSO Scholarship.