Heteroscedastic Uncertainty Estimation Framework for Unsupervised Registration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhang, Xiaoran, Pak, Daniel H., Ahn, Shawn S., Li, Xiaoxiao, You, Chenyu, Staib, Lawrence H., Sinusas, Albert J., Wong, Alex, Duncan, James S.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929425557225472
author Zhang, Xiaoran
Pak, Daniel H.
Ahn, Shawn S.
Li, Xiaoxiao
You, Chenyu
Staib, Lawrence H.
Sinusas, Albert J.
Wong, Alex
Duncan, James S.
author_facet Zhang, Xiaoran
Pak, Daniel H.
Ahn, Shawn S.
Li, Xiaoxiao
You, Chenyu
Staib, Lawrence H.
Sinusas, Albert J.
Wong, Alex
Duncan, James S.
contents Deep learning methods for unsupervised registration often rely on objectives that assume a uniform noise level across the spatial domain (e.g. mean-squared error loss), but noise distributions are often heteroscedastic and input-dependent in real-world medical images. Thus, this assumption often leads to degradation in registration performance, mainly due to the undesired influence of noise-induced outliers. To mitigate this, we propose a framework for heteroscedastic image uncertainty estimation that can adaptively reduce the influence of regions with high uncertainty during unsupervised registration. The framework consists of a collaborative training strategy for the displacement and variance estimators, and a novel image fidelity weighting scheme utilizing signal-to-noise ratios. Our approach prevents the model from being driven away by spurious gradients caused by the simplified homoscedastic assumption, leading to more accurate displacement estimation. To illustrate its versatility and effectiveness, we tested our framework on two representative registration architectures across three medical image datasets. Our method consistently outperforms baselines and produces sensible uncertainty estimates. The code is publicly available at \url{https://voldemort108x.github.io/hetero_uncertainty/}.
format Preprint
id arxiv_https___arxiv_org_abs_2312_00836
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Heteroscedastic Uncertainty Estimation Framework for Unsupervised Registration
Zhang, Xiaoran
Pak, Daniel H.
Ahn, Shawn S.
Li, Xiaoxiao
You, Chenyu
Staib, Lawrence H.
Sinusas, Albert J.
Wong, Alex
Duncan, James S.
Image and Video Processing
Computer Vision and Pattern Recognition
Deep learning methods for unsupervised registration often rely on objectives that assume a uniform noise level across the spatial domain (e.g. mean-squared error loss), but noise distributions are often heteroscedastic and input-dependent in real-world medical images. Thus, this assumption often leads to degradation in registration performance, mainly due to the undesired influence of noise-induced outliers. To mitigate this, we propose a framework for heteroscedastic image uncertainty estimation that can adaptively reduce the influence of regions with high uncertainty during unsupervised registration. The framework consists of a collaborative training strategy for the displacement and variance estimators, and a novel image fidelity weighting scheme utilizing signal-to-noise ratios. Our approach prevents the model from being driven away by spurious gradients caused by the simplified homoscedastic assumption, leading to more accurate displacement estimation. To illustrate its versatility and effectiveness, we tested our framework on two representative registration architectures across three medical image datasets. Our method consistently outperforms baselines and produces sensible uncertainty estimates. The code is publicly available at \url{https://voldemort108x.github.io/hetero_uncertainty/}.
title Heteroscedastic Uncertainty Estimation Framework for Unsupervised Registration
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2312.00836