目的 弥散加权成像（Diffusion weight imaging ,DWI）作为一种新型医学影像成像技术已逐渐成为诊断心脏、大脑、肾脏、肝脏等器官中神经、纤维组织病变的重要方法和手段。与传统的MRI成像相比，DWI图像通过使用不同的扩散方向矢量，在不同的扩散参数下，DWI图像呈现的灰度信息也有所不同。目前尚无相关文献提出针对DWI图像版权信息进行有效的保护的相关研究。方法 为有效的保护病人的DWI图像版权信息，本文提出一种基于DWI图像的整数小波变换域（Integer wavelet transform，IWT）统计直方图鲁棒水印算法。该算法首先通过最大类间方差分割算法和面积控制阈值获取指定断层中带有弥散梯度方向图像的前景区域作为待嵌入区域，对待嵌入区域使用整数小波变换获取低频子带系数，利用固定步长对低频子带系数直方图相邻簇的比值关系用于水印嵌入；最后提出DWI表观系数与弥散张量成像（Diffusion tensor imaging, DTI）中弥散张量值的可逆关系设计成为密钥，利用该密钥将嵌入水印后的DWI图像再次加密，从而有效保护DWI图像的版权信息。结论 实验表明该算法引入的水印信息对DWI图像中的纤维参数改变量极小。在各项同性和纤维方向改变个数上，本文算法在文献方法中分别降低了100和30多；在可视质量上，本文提高约8dB。在高斯噪声、小角度旋转等攻击中，本文算法能够提供较高的提取水印准确率。结果：对医生诊断的影响在可接受的范围内，且在感兴趣区域对各种常见攻击，具有较高的安全性和鲁棒性。
Research on robust watermarking algorithm of dispersion weighted image
Chen Yi,Li Zhi,Wang Lihui,Zhang Jian,Wang Guomei(Key Laboratory Of Intelligent Medical Image Analysis and Precise Diagnosis of Guizhou Province)
Objective Diffusion weight imaging (DWI), as a new medical imaging technique, transforms the diffusion motion of water molecules in tissues into grayscale or other parameters information of the image by applying multi-directional diffusion magnetic sensitive gradients under each diffusion sensitive gradient. This technique can be used for the auxiliary diagnosis of living heart myocardial fiber modeling, brain fiber, lesions of the central nervous system, liver fibrosis and other diseases. With the popularization of telemedicine diagnostic technology, more and more DWI data are used for remote diagnosis and scientific research. DWI images are originally stored and used on a single machine in the hospital need to be transmitted and used over the network. Many scholars have proposed many watermarking algorithms for protecting medical images, such as reversible watermarking algorithm, robust reversible watermarking algorithm, and zero-watermarking algorithm. The advantage of the reversible algorithm is that can be completely nondestructive image recovery. The robustness of the reversible watermarking algorithm is too weak to guarantee the existence of reversible watermark when the embedded images are attacked by some intentional or unintentional attacks. Therefore, some researchers proposed the robust reversible watermarking algorithm. The robust reversible watermarking algorithm which could restore the original picture when there is no attacked and could draw the embedded watermarking. It is used to ensure its robust reversible performance should be carried additional information, so it needs to consume more transmission bandwidth. For some robust reversible watermarks are constructed by dual watermarking, they depend on each other information to extract the watermarking. In order to protect medical images by other methods, some researchers contribute zero-watermarking algorithm which is different from the traditional method which must embed information into images. It can gain the internal feature in the data to build the binary watermarking, and then we save it in the third party. When the image is been used by other people who do not get the license, we could use the zero-watermarking to prove copyright. Thus, the zero-watermarking algorithm, as a non-embedded algorithm, can not actively complete the protection of property information. If we want to ensure that the medical image watermarking information has certain robustness, the robust watermarking algorithm plays an irreplaceable role. In order to prevent unauthorized DWI images from being used or tampered, a robust watermarking algorithm based on DWI image is proposed in this paper. Methods The algorithm firstly obtains specified slips by the maximum inter-class variance segmentation algorithm and area control threshold which set the threshold to ensure that the selected slice has a sufficiently embedding area. Because the tip and the bottom of the heart are not suitable for embedding. The foreground region with diffusion gradient direction image in the selected as the region which is prepared to embed. We obtain the low-frequency sub-band coefficient by using integer wavelet transform (IWT) in the default region. Then we count the low frequency and the low sub-frequency coefficient is analyzed by using the fixed step length which follows the characteristics of the coefficient of DWI images. The ratio relation of adjacent clusters in the histogram subject area is adjusted for watermark embedding. Finally, we proposed to design the quantitative reversible relationship between DWI apparent coefficient and diffusion tensor imaging (DTI) as the key and use this key to encrypt the DWI image after embedding watermark again, it could be effectively protecting the copyright information of DWI image. Result In this paper, the algorithm can maintain the robustness of the algorithm and reduce the change of DTI parameters in the experiment of robustness and the change of the parameters of DTI after embedding. It has excellent robustness in attack experiments such as Gaussian noise, contrast expansion, and small angle rotation. In the experiment of parameter change measurement before and after embedding, the algorithm is greatly reduced in the change of the isotropic and fiber main direction of the myocardial fiber. In our proposed method, the average change of the isotropic is reduced more than 100 and the main direction of the fiber is reduced more than 30 in the same database. In the visual quality of the algorithm, The Peak Signal Noise Ratio (PSNR) is about 8dB higher than the comparative literature. Conclusion In this paper, an embedded selection feedback mechanism is firstly proposed to carry out the selection of watermark embedding according to the actual demand in embedding. Then, the statistical histogram of the sub-band coefficient is constructed by specifying the fixed step length according to the characteristics of the wavelet transform coefficient of the DWI image. Finally, the reversible key algorithm based on the quantitative relationship between DWI and DTI is constructed. Experiments show that this algorithm can be applied to watermark embedding of dispersion weighted imaging and can satisfy the fiber direction as little as possible.