Non-rigid medical image registration is a popular subject in the research areas of the medical image and has an important clinical value. In this paper we put forward an improved algorithm of Demons, together with the conservation of gray model and local structure tensor conservation model, to construct a new energy function processing multi-modal registration problem. We then applied the L-BFGS algorithm to optimize the energy function and solve complex three-dimensional data optimization problem. And finally we used the multi-scale hierarchical refinement ideas to solve large deformation registration. The experimental results showed that the proposed algorithm for large deformation and multi-modal three-dimensional medical image registration had good effects.
The medical image registration between preoperative three-dimensional (3D) scan data and intraoperative two-dimensional (2D) image is a key technology in the surgical navigation. Most previous methods need to generate 2D digitally reconstructed radiographs (DRR) images from the 3D scan volume data, then use conventional image similarity function for comparison. This procedure includes a large amount of calculation and is difficult to archive real-time processing. In this paper, with using geometric feature and image density mixed characteristics, we proposed a new similarity measure function for fast 2D-3D registration of preoperative CT and intraoperative X-ray images. This algorithm is easy to implement, and the calculation process is very short, while the resulting registration accuracy can meet the clinical use. In addition, the entire calculation process is very suitable for highly parallel numerical calculation by using the algorithm based on CUDA hardware acceleration to satisfy the requirement of real-time application in surgery.
Medical image registration is very challenging due to the various imaging modality, image quality, wide inter-patients variability, and intra-patient variability with disease progressing of medical images, with strict requirement for robustness. Inspired by semantic model, especially the recent tremendous progress in computer vision tasks under bag-of-visual-word framework, we set up a novel semantic model to match medical images. Since most of medical images have poor contrast, small dynamic range, and involving only intensities and so on, the traditional visual word models do not perform very well. To benefit from the advantages from the relative works, we proposed a novel visual word model named directional visual words, which performs better on medical images. Then we applied this model to do medical registration. In our experiment, the critical anatomical structures were first manually specified by experts. Then we adopted the directional visual word, the strategy of spatial pyramid searching from coarse to fine, and the k-means algorithm to help us locating the positions of the key structures accurately. Sequentially, we shall register corresponding images by the areas around these positions. The results of the experiments which were performed on real cardiac images showed that our method could achieve high registration accuracy in some specific areas.
With the development of image-guided surgery and radiotherapy, the demand for medical image registration is stronger and the challenge is greater. In recent years, deep learning, especially deep convolution neural networks, has made excellent achievements in medical image processing, and its research in registration has developed rapidly. In this paper, the research progress of medical image registration based on deep learning at home and abroad is reviewed according to the category of technical methods, which include similarity measurement with an iterative optimization strategy, direct estimation of transform parameters, etc. Then, the challenge of deep learning in medical image registration is analyzed, and the possible solutions and open research are proposed.