Deformable Registration of Digital Images

This paper proposes a novel elastic model and presents a deformable registration method based on the model. The method registers images without the need to extract features from the images, and therefore works directly on grey-level images. A new similarity metric is given on which the formation of external forces is based. The registration method, taking the coarse-to-fine strategy constructs external forces in larger scales for the first few iterations to rely more on global evidence, and then in smaller scales for later iterations to allow local relinements. The stiffness of the elastic body decreases as the process proceeds.To make it widely applicable, the method is not restricted to any trpe of transformation. The variations between images are thought as general free-form deformations.Because the elastic model designed is linearized, it can be solved very efficiently with high accuracy.The method has been successfully tested on MRI images. It will certainly find other uses such as matching time-varying sequences of pictures for moion analysis,fitting templates into images for non-rigid object recognition, maching stereo images for shape recovery etc.

Optical Flow with Learning Feature for Deformable Medical Image Registration

Deformable medical image registration plays a vital role in medical image applications,such as placing different temporal images at the same time point or different modality images into the same coordinate system.Various strategies have been developed to satisfy the increasing needs of deformable medical image registration.One popular registration method is estimating the displacement field by computing the optical flow between two images.The motion field(flow field)is computed based on either gray-value or handcrafted descriptors such as the scale-invariant feature transform(SIFT).These methods assume that illumination is constant between images.However,medical images may not always satisfy this assumption.In this study,we propose a metric learning-based motion estimation method called Siamese Flow for deformable medical image registration.We train metric learners using a Siamese network,which produces an image patch descriptor that guarantees a smaller feature distance in two similar anatomical structures and a larger feature distance in two dissimilar anatomical structures.In the proposed registration framework,the flow field is computed based on such features and is close to the real deformation field due to the excellent feature representation ability of the Siamese network.Experimental results demonstrate that the proposed method outperforms the Demons,SIFT Flow,Elastix,and VoxelMorph networks regarding registration accuracy and robustness,particularly with large deformations.

GAN-DIRNet:A Novel Deformable Image Registration Approach for Multimodal Histological Images

Multi-modal histological image registration tasks pose significant challenges due to tissue staining operations causing partial loss and folding of tissue.Convolutional neural network(CNN)and generative adversarial network(GAN)are pivotal inmedical image registration.However,existing methods often struggle with severe interference and deformation,as seen in histological images of conditions like Cushing’s disease.We argue that the failure of current approaches lies in underutilizing the feature extraction capability of the discriminator inGAN.In this study,we propose a novel multi-modal registration approach GAN-DIRNet based on GAN for deformable histological image registration.To begin with,the discriminators of two GANs are embedded as a new dual parallel feature extraction module into the unsupervised registration networks,characterized by implicitly extracting feature descriptors of specific modalities.Additionally,modal feature description layers and registration layers collaborate in unsupervised optimization,facilitating faster convergence and more precise results.Lastly,experiments and evaluations were conducted on the registration of the Mixed National Institute of Standards and Technology database(MNIST),eight publicly available datasets of histological sections and the Clustering-Registration-Classification-Segmentation(CRCS)dataset on the Cushing’s disease.Experimental results demonstrate that our proposed GAN-DIRNet method surpasses existing approaches like DIRNet in terms of both registration accuracy and time efficiency,while also exhibiting robustness across different image types.

ASRNet: Adversarial Segmentation and Registration Networks for Multispectral Fundus Images

Multispectral imaging (MSI) technique is often used to capture imagesof the fundus by illuminating it with different wavelengths of light. However,these images are taken at different points in time such that eyeball movementscan cause misalignment between consecutive images. The multispectral imagesequence reveals important information in the form of retinal and choroidal bloodvessel maps, which can help ophthalmologists to analyze the morphology of theseblood vessels in detail. This in turn can lead to a high diagnostic accuracy of several diseases. In this paper, we propose a novel semi-supervised end-to-end deeplearning framework called “Adversarial Segmentation and Registration Nets”(ASRNet) for the simultaneous estimation of the blood vessel segmentation andthe registration of multispectral images via an adversarial learning process. ASRNet consists of two subnetworks: (i) A segmentation module S that fulfills theblood vessel segmentation task, and (ii) A registration module R that estimatesthe spatial correspondence of an image pair. Based on the segmention-drivenregistration network, we train the segmentation network using a semi-supervisedadversarial learning strategy. Our experimental results show that the proposedASRNet can achieve state-of-the-art accuracy in segmentation and registrationtasks performed with real MSI datasets.

A review on 3D deformable image registration and its application in dose warping

Deformable image registration(DIR)has been well explored in recent decades,and it is widely utilized in clinical tasks,especially dose warping.Nowadays,as deep learning(DL)develops rapidly,many DL-based methods were also applied in DIR.This paper reviews DL-based DIR methods in recent years and the application of DIR in dose warping.We collected and categorized the latest DL-based DIR studies.A thorough review of each category was presented,in which studies were discussed based on their supervision,advantage,and challenges.Then,we reviewed DIR-based dose warping and discussed its rationale,feasibility,successes,and difficulties.Lastly,we summarized the review on both parts and discussed their future development trend.

Multi-affine registration using local polynomial expansion

In this paper, we present a non-linear (multi-affine) registration algorithm based on a local polynomial expansion model. We generalize previous work using a quadratic polynomial expansion model. Local affine models are estimated using this generalized model analytically and iteratively, and combined to a deformable registration algorithm. Experiments show that the affine parameter calculations derived from this quadratic model are more accurate than using a linear model. Experiments further indicate that the multi-affine deformable registration method can handle complex non-linear deformation fields necessary for deformable registration, and a faster convergent rate is verified from our comparison experiment.