Hybrid Atlas Building with Deep Registration Priors

TL;DR

This paper presents a hybrid atlas building method that leverages deep learning registration priors to significantly reduce computational costs while maintaining high-quality results in 3D brain MRI analysis.

Contribution

The paper introduces a novel hybrid framework combining traditional registration with learned deep registration priors for efficient atlas construction.

Findings

Reduces computational cost of atlas building

Maintains high-quality registration results

Flexible integration of various deep learning registration methods

Abstract

Registration-based atlas building often poses computational challenges in high-dimensional image spaces. In this paper, we introduce a novel hybrid atlas building algorithm that fast estimates atlas from large-scale image datasets with much reduced computational cost. In contrast to previous approaches that iteratively perform registration tasks between an estimated atlas and individual images, we propose to use learned priors of registration from pre-trained neural networks. This newly developed hybrid framework features several advantages of (i) providing an efficient way of atlas building without losing the quality of results, and (ii) offering flexibility in utilizing a wide variety of deep learning based registration methods. We demonstrate the effectiveness of this proposed model on 3D brain magnetic resonance imaging (MRI) scans.