WSSAMNet: Weakly Supervised Semantic Attentive Medical Image Registration Network

TL;DR

WSSAMNet is a weakly supervised medical image registration network that uses segmentation masks to improve alignment, showing competitive results on BraTSReg challenge data.

Contribution

It introduces a two-step weakly supervised approach combining segmentation and registration for medical images, enhancing registration accuracy.

Findings

Competitive performance on BraTSReg data

Effective use of segmentation masks for registration

Outperforms some existing methods

Abstract

We present WSSAMNet, a weakly supervised method for medical image registration. Ours is a two step method, with the first step being the computation of segmentation masks of the fixed and moving volumes. These masks are then used to attend to the input volume, which are then provided as inputs to a registration network in the second step. The registration network computes the deformation field to perform the alignment between the fixed and the moving volumes. We study the effectiveness of our technique on the BraTSReg challenge data against ANTs and VoxelMorph, where we demonstrate that our method performs competitively.