Brain Tumor Sequence Registration with Non-iterative Coarse-to-fine Networks and Dual Deep Supervision

TL;DR

This paper presents a novel deep learning-based non-iterative coarse-to-fine network with dual supervision for brain tumor sequence registration, effectively handling large deformations and missing correspondences, achieving competitive results in BraTS-Reg 2022.

Contribution

The study introduces NICE-Net with dual deep supervision, enhancing brain tumor registration accuracy for large deformations and missing data in MRI scans.

Findings

Achieved mean absolute error of 3.387 on validation set.

Placed 4th in BraTS-Reg 2022 testing phase.

Demonstrated effectiveness of dual supervision in registration accuracy.

Abstract

In this study, we focus on brain tumor sequence registration between pre-operative and follow-up Magnetic Resonance Imaging (MRI) scans of brain glioma patients, in the context of Brain Tumor Sequence Registration challenge (BraTS-Reg 2022). Brain tumor registration is a fundamental requirement in brain image analysis for quantifying tumor changes. This is a challenging task due to large deformations and missing correspondences between pre-operative and follow-up scans. For this task, we adopt our recently proposed Non-Iterative Coarse-to-finE registration Networks (NICE-Net) - a deep learning-based method for coarse-to-fine registering images with large deformations. To overcome missing correspondences, we extend the NICE-Net by introducing dual deep supervision, where a deep self-supervised loss based on image similarity and a deep weakly-supervised loss based on manually annotated landmarks are deeply embedded into the NICE-Net. At the BraTS-Reg 2022, our method achieved a competitive result on the validation set (mean absolute error: 3.387) and placed 4th in the final testing phase (Score: 0.3544).