Blind Integrity Verification of Medical Images

• Published in: IEEE transactions on information technology in biomedicine Vol. 16; no. 6; pp. 1122 - 1126
• Main Authors: Hui Huang, Coatrieux, G., Huazhong Shu, Limin Luo, Roux, C.
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.11.2012; Institute of Electrical and Electronics Engineers
• Subjects: Bioengineering; Blind forensics; Blinds; Blocking; classification; Computer Security; Databases, Factual; Diagnostic Imaging - standards; Discrete cosine transforms; Feature extraction; Filtering; Forensic engineering; Forensics; Histograms; Humans; image moments; Image Processing, Computer-Assisted - methods; Image Processing, Computer-Assisted - standards; Life Sciences; Medical; Medical diagnostic imaging; Medical Informatics Applications; Reproducibility of Results; Statistics; Support Vector Machine; Support vector machines
• ISSN: 1089-7771; 1558-0032
• DOI: 10.1109/TITB.2012.2207435

This paper presents the first method of digital blind forensics within the medical imaging field with the objective to detect whether an image has been modified by some processing (e.g., filtering, lossy compression, and so on). It compares two image features: the histogram statistics of reorganized block-based discrete cosine transform coefficients, originally proposed for steganalysis purposes, and the histogram statistics of reorganized block-based Tchebichef moments. Both features serve as input of a set of support vector machine classifiers built in order to discriminate tampered images from original ones as well as to identify the nature of the global modification one image may have undergone. Performance evaluation, conducted in application to different medical image modalities, shows that these image features can help, independently or jointly, to blindly distinguish image processing or modifications with a detection rate greater than 70%. They also underline the complementarity of these features.