Bayesian Fully Convolutional Networks for Brain Image Registration

Abstract

The purpose of medical image registration is to find geometric transformations that align two medical images so that the corresponding voxels on two images are spatially consistent. Nonrigid medical image registration is a key step in medical image processing, such as image comparison, data fusion, target recognition, and pathological change analysis. Existing registration methods only consider registration accuracy but largely neglect the uncertainty of registration results. In this work, a method based on the Bayesian fully convolutional neural network is proposed for nonrigid medical image registration. The proposed method can generate a geometric uncertainty map to calculate the uncertainty of registration results. This uncertainty can be interpreted as a confidence interval, which is essential for judging whether the source data are abnormal. Moreover, the proposed method introduces group normalization, which is conducive to the network convergence of the Bayesian neural network. Some representative learning-based image registration methods are compared with the proposed method on different image datasets. Experimental results show that the registration accuracy of the proposed method is better than that of the methods, and its antifolding performance is comparable to that of fast image registration and VoxelMorph. Furthermore, the proposed method can evaluate the uncertainty of registration results.

An optimal self adaptive deep neural network and spine‐kernelled chirplet transform for image registration

“…Total training loss is defined as the sum of dissimilar images and regularization, 27 and it is expressed below equation, $$$$ {A}_{to}={A}_{im}\left(R,V\right)+\beta {W}_1(r)+\alpha {W}_2(r) $$$$ The total training loss with cross‐correlation (CC) is referred to as similarity between the target image and the warped source image is given below, $$$$ CC\left(b,t\circ \varphi \right)=\frac{{\left({\sum}_{y\in \sigma}\left(b(y)-\overrightarrow{b(y)}\right)\left(t\circ \varphi (y)-\overline{t\circ \varphi }(y)\right)\right)}^2}{\left({\sum}_{y\in \sigma }{\left(b(y)-c\right)}^2\right)\left({\sum}_{y\in \sigma }{\left(t\circ \varphi (y)-\overline{t\circ \varphi }(y)\right)}^2\right)} $$$$ where, $$$ b(y) $$$, $$$ b...$…”

“…Total training loss is defined as the sum of dissimilar images and regularization, 27 and it is expressed below equation,…”

Remote Sensing and Deep Learning to Understand Noisy OpenStreetMap

“…The suggested image registration method employs an area-based approach, to achieve similarity using the Cross-Correlation (CCR) measure. We use a CCR-based technique for resolving the building misalignment error [31]. CCR has been used to solve registration issues in a variety of domains with great success [32,33].…”

A Review on Bayesian Deep Learning in Healthcare: Applications and Challenges

“…The paper is one of the few works that have deployed MCMC for Bayesian sampling. Cui et al [73] used Bayesian CNN for brain image registration. The suggested method used MC-dropout as a Bayesian method to sample from posterior for the registration task as well as producing the geometric uncertainty map for the uncertainty associated with the registration process.…”