FlowReg: Fast Deformable Unsupervised Medical Image Registration using Optical Flow (2021-01-24T00:00:00.000000Z)

TL;DR

Experimental results show that FlowReg is able to obtain high intensity and spatial similarity between the moving and the fixed volumes while maintaining the shape and structure of anatomy and pathology.

Abstract

In this work we propose FlowReg, a deep learning-based framework that performs unsupervised image registration for neuroimaging applications. The system is composed of two architectures that are trained sequentially: FlowReg-A which affinely corrects for gross differences between moving and fixed volumes in 3D followed by FlowReg-O which performs pixel-wise deformations on a slice-by-slice basis for fine tuning in 2D. FlowReg-A warps the moving volume using gross global parameters to align rotation, scale, shear, and translation to the fixed volume. A correlation loss that encourages global alignment between the moving and the fixed volumes is employed to regress the affine parameters. The deformable network FlowReg-O operates on 2D image slices and is based on the optical flow CNN network that is adapted to neuroimaging with three loss components. The photometric loss minimizes pixel intensity differences, the smoothness loss encourages similar magnitudes between neighbouring vectors, and a correlation loss that is used to maintain the intensity similarity between fixed and moving image slices. The proposed method is compared to four open source registration techniques ANTs, Demons, SE, and Voxelmorph for FLAIR MRI applications. In total, 4643 FLAIR MR imaging volumes (approximately 255,000 image slices) are used from dementia and vascular disease cohorts, acquired from over 60 international centres with varying acquisition parameters. To quantitatively assess the performance of the registration tools, a battery of novel validation metrics are proposed that focus on the structural integrity of tissues, spatial alignment, and intensity similarity. Experimental results show FlowReg (FlowReg-A+O) performs better than iterative-based registration algorithms for intensity and spatial alignment metrics with a Pixelwise Agreement (PWA) of 0.65, correlation coefficient (R) of 0.80, and Mutual Information (MI) of 0.29. Among the deep learning frameworks evaluated, FlowReg-A or FlowReg-A+O provided the highest performance over all but one of the metrics. Results show that FlowReg is able to obtain high intensity and spatial similarity between the moving and the fixed volumes while maintaining the shape and structure of anatomy and pathology.

References58 items

Cerebrovascular pathology in Alzheimer's disease: Hopes and gaps

Unsupervised 3D End-to-end Deformable Network for Brain MRI Registration

Whole Volume Brain Extraction for Multi-Centre, Multi-Disease FLAIR MRI Datasets.

White matter hyperintensities in vascular contributions to cognitive impairment and dementia (VCID): Knowledge gaps and opportunities

Characterization of White Matter Hyperintensities in Large-Scale MRI-Studies

Unsupervised 3D End-to-End Medical Image Registration With Volume Tweening Network

An Efficient Automatic Midsagittal Plane Extraction in Brain MRI

A deep learning framework for unsupervised affine and deformable image registration

Adversarial Similarity Network for Evaluating Image Alignment in Deep Learning Based Registration

Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration

An Unsupervised Learning Model for Deformable Medical Image Registration

Training CNNs for Image Registration from Few Samples with Model-based Data Augmentation

Deformable Image Registration Based on Similarity-Steered CNN Regression

Nonrigid Image Registration Using Multi-scale 3D Convolutional Neural Networks

Classification and characterization of periventricular and deep white matter hyperintensities on MRI: A study in older adults

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

White Matter Diseases with Radiologic-Pathologic Correlation.

Back to Basics: Unsupervised Learning of Optical Flow via Brightness Constancy and Motion Smoothness

SimpleElastix: A User-Friendly, Multi-lingual Library for Medical Image Registration

Unsupervised CNN for Single View Depth Estimation: Geometry to the Rescue

Current trends in medical image registration and fusion

Spatial Transformer Networks

U-Net: Convolutional Networks for Biomedical Image Segmentation

FlowNet: Learning Optical Flow with Convolutional Networks

Classifying Dementia Using Local Binary Patterns from Different Regions in Magnetic Resonance Images

Adam: A Method for Stochastic Optimization

Multi-atlas segmentation of biomedical images: A survey

Improving Atlas-Based Medical Image Segmentation with a Relaxed Object Search

Neuroimaging standards for research into small vessel disease and its contribution to ageing and neurodegeneration

Atherosclerosis imaging and the Canadian Atherosclerosis Imaging Network.

Survey of Medical Image Registration

Longitudinal Image Registration with Non-uniform Appearance Change

Robust White Matter Lesion Segmentation in FLAIR MRI

The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis

A robust method to estimate the intracranial volume across MRI field strengths (1.5T and 3T)

Nonrigid registration of multiple sclerosis brain images using lesion inpainting for morphometry or lesion mapping

Classification of White Matter Lesions on Magnetic Resonance Imaging in Elderly Persons

Symmetric diffeomorphic image registration with cross-correlation: Evaluating automated labeling of elderly and neurodegenerative brain

Contribution of Diffusion-Weighted Imaging in the Evaluation of Diffuse White Matter Ischemic Lesions in Fetuses: Correlations with Fetopathologic Findings

Neuroimaging of White Matter in Aging and Dementia

Imaging in multiple sclerosis

Comparing white matter lesions on T2 and FLAIR MRI in the Sydney Older Persons Study

Non-rigid Atlas to Subject Registration with Pathologies for Conformal Brain Radiotherapy

White Matter Lesions and Dementia MRI‐Pathological Correlation

Apoptosis of astrocytes with enhanced lysosomal activity and oligodendrocytes in white matter lesions in Alzheimer's disease

Robust midsagittal plane extraction from normal and pathological 3-D neuroradiology images

Understanding the "Demon's Algorithm": 3D Non-rigid Registration by Gradient Descent

Image matching as a diffusion process: an analogy with Maxwell's demons

Multimodality image registration by maximization of mutual information

Clinical significance of MRI white matter lesions in the elderly

Determining Optical Flow

TensorFlow: Large-scale machine learning on heterogeneous systems

Ibáñez, and The Insight Software Consortium. The ITK Software Guide

AHA / ASA Scientific Statement Vascular Contributions to Cognitive Impairment and Dementia

elastix: A Toolbox for Intensity-Based Medical Image Registration

Fast Non-Rigid Matching of 3D Medical Images

Flair templates-brainder

Computer-assisted Segmentation of White Matter Lesions in 3d Mr Images Using Support Vector Machine 1