NPBDREG: A Non-parametric Bayesian Deep-Learning Based Approach for Diffeomorphic Brain MRI Registration

08/15/2021
by   Samah Khawaled, et al.
20

Quantification of uncertainty in deep-neural-networks (DNN) based image registration algorithms plays an important role in the safe deployment of real-world medical applications and research-oriented processing pipelines, and in improving generalization capabilities. Currently available approaches for uncertainty estimation, including the variational encoder-decoder architecture and the inference-time dropout approach, require specific network architectures and assume parametric distribution of the latent space which may result in sub-optimal characterization of the posterior distribution for the predicted deformation-fields. We introduce the NPBDREG, a fully non-parametric Bayesian framework for unsupervised DNN-based deformable image registration by combining an optimizer with stochastic gradient Langevin dynamics (SGLD) to characterize the true posterior distribution through posterior sampling. The NPBDREG provides a principled non-parametric way to characterize the true posterior distribution, thus providing improved uncertainty estimates and confidence measures in a theoretically well-founded and computationally efficient way. We demonstrated the added-value of NPBDREG, compared to the baseline probabilistic unsupervised model (PrVXM), on brain MRI images registration using 390 image pairs from four publicly available databases: MGH10, CMUC12, ISBR18 and LPBA40. The NPBDREG shows a slight improvement in the registration accuracy compared to PrVXM (Dice score of 0.73 vs. 0.68, p ≪ 0.01), a better generalization capability for data corrupted by a mixed structure noise (e.g Dice score of 0.729 vs. 0.686 for α=0.2) and last but foremost, a significantly better correlation of the predicted uncertainty with out-of-distribution data (r>0.95 vs. r<0.5).

READ FULL TEXT

page 2

page 7

page 8

page 9

research
08/10/2020

Unsupervised Deep-Learning Based Deformable Image Registration: A Bayesian Framework

Unsupervised deep-learning (DL) models were recently proposed for deform...
research
08/08/2022

NPB-REC: Non-parametric Assessment of Uncertainty in Deep-learning-based MRI Reconstruction from Undersampled Data

Uncertainty quantification in deep-learning (DL) based image reconstruct...
research
01/12/2017

Probabilistic Diffeomorphic Registration: Representing Uncertainty

This paper presents a novel mathematical framework for representing unce...
research
11/22/2021

Deformable Image Registration with Deep Network Priors: a Study on Longitudinal PET Images

Longitudinal image registration is challenging and has not yet benefited...
research
04/07/2016

Reinterpreting the Transformation Posterior in Probabilistic Image Registration

Probabilistic image registration methods estimate the posterior distribu...
research
07/02/2019

A Closest Point Proposal for MCMC-based Probabilistic Surface Registration

In this paper, we propose a non-rigid surface registration algorithm tha...
research
07/22/2019

Bayesian estimation of probabilistic sensitivity measures

Computer experiments are becoming increasingly important in scientific i...

Please sign up or login with your details

Forgot password? Click here to reset