Published on Tue Nov 03 2020

Uncertainty Estimation in Medical Image Localization: Towards Robust Anterior Thalamus Targeting for Deep Brain Stimulation

Han Liu, Can Cui, Dario J. Englot, Benoit M. Dawant

Atlas-based methods are the standard approaches for automatic targeting of the Anterior Nucleus of the Thalamus (ANT) for Deep Brain Stimulation. These are known to lack robustness when anatomic differences between subjects are large. We propose a novel two-stage deep learning (DL) framework.

0
0
0
Abstract

Atlas-based methods are the standard approaches for automatic targeting of the Anterior Nucleus of the Thalamus (ANT) for Deep Brain Stimulation (DBS), but these are known to lack robustness when anatomic differences between atlases and subjects are large. To improve the localization robustness, we propose a novel two-stage deep learning (DL) framework, where the first stage identifies and crops the thalamus regions from the whole brain MRI and the second stage performs per-voxel regression on the cropped volume to localize the targets at the finest resolution scale. To address the issue of data scarcity, we train the models with the pseudo labels which are created based on the available labeled data using multi-atlas registration. To assess the performance of the proposed framework, we validate two sampling-based uncertainty estimation techniques namely Monte Carlo Dropout (MCDO) and Test-Time Augmentation (TTA) on the second-stage localization network. Moreover, we propose a novel uncertainty estimation metric called maximum activation dispersion (MAD) to estimate the image-wise uncertainty for localization tasks. Our results show that the proposed method achieved more robust localization performance than the traditional multi-atlas method and TTA could further improve the robustness. Moreover, the epistemic and hybrid uncertainty estimated by MAD could be used to detect the unreliable localizations and the magnitude of the uncertainty estimated by MAD could reflect the degree of unreliability for the rejected predictions.

Fri Jun 01 2018
Computer Vision
Spatially Localized Atlas Network Tiles Enables 3D Whole Brain Segmentation from Limited Data
Historically,multi-atlas segmentation (MAS) has been regarded as the de facto standard method for whole brain segmentation. In this paper, we propose the spatially localized atlas network tiles (SLANT) method to distribute multiple independent 3D fully convolutional networks. This strategy simplifies the whole brain learning task to localized sub-tasks.
0
0
0
Thu Mar 28 2019
Computer Vision
3D Whole Brain Segmentation using Spatially Localized Atlas Network Tiles
Detailed whole brain segmentation is an essential quantitative technique. We propose the spatially localized atlas network tiles (SLANT) method to distribute multiple independent 3D fully convolutional networks. The method achieved superior performance compared with multi-atlas segmentation methods, while reducing the computational time from >30 hours to 15 minutes.
0
0
0
Mon Jul 01 2019
Computer Vision
Multi-atlas image registration of clinical data with automated quality assessment using ventricle segmentation
The method consists of automatically segmenting the ventricles of a given scan using a neural network. The automated ventricle segmentation reached a Dice coefficient with manual annotations of 0.89 in the single-site dataset and 0.83 in the multi-center dataset.
0
0
0
Tue Aug 13 2019
Computer Vision
Few Labeled Atlases are Necessary for Deep-Learning-Based Segmentation
biomedical image segmentation in the scenario of only a few labeled MR images. Current multi-atlas based segmentation methods use image registration to warp segments onto a new scan.
0
0
0
Mon Jan 07 2019
Computer Vision
Reproducibility Evaluation of SLANT Whole Brain Segmentation Across Clinical Magnetic Resonance Imaging Protocols
Whole brain segmentation on structural magnetic resonance imaging (MRI) is essential for understanding neuroanatomical-functional relationships. In past few years, deep convolutional neural network (DCNN) segmentation methods have demonstrated their advantages in accuracy and computational efficiency. Recently, we proposed
0
0
0
Sun Aug 15 2021
Computer Vision
NPBDREG: A Non-parametric Bayesian Deep-Learning Based Approach for Diffeomorphic Brain MRI Registration
NPBDREG is a fully non-parametric Bayesian framework for unsupervised DNN-based deformable image registration. It combines Adam optimizer with stochastic gradient Langevin dynamics (SGLD) to characterize the true posterior distribution through posterior sampling. The NPBDREG shows a slight improvement in the registration accuracy compared to PrVXM.
3
1
0