Decoder-Only Image Registration

In unsupervised medical image registration, encoder-decoder architectures are widely used to predict dense, full-resolution displacement fields from paired images. Despite their popularity, we question the necessity of making both the encoder and decoder learnable. To address this, we propose LessNe...

Full description

Saved in:
Bibliographic Details
Published inIEEE transactions on medical imaging Vol. 44; no. 8; pp. 3356 - 3369
Main Authors Jia, Xi, Lu, Wenqi, Cheng, Xinxing, Duan, Jinming
Format Journal Article
LanguageEnglish
Published United States IEEE 01.08.2025
Subjects
Accuracy
Algorithms
Brain - diagnostic imaging
Computer architecture
Databases, Factual
Decoder-only
Decoding
Deformation
diffeomorphic
efficient
Heart - diagnostic imaging
Humans
Image registration
Imaging, Three-Dimensional - methods
Iterative methods
Magnetic Resonance Imaging - methods
Medical diagnostic imaging
Optimization
Training
Transformers
U-Net
Online AccessGet full text

Cover

More Information
Summary:In unsupervised medical image registration, encoder-decoder architectures are widely used to predict dense, full-resolution displacement fields from paired images. Despite their popularity, we question the necessity of making both the encoder and decoder learnable. To address this, we propose LessNet, a simplified network architecture with only a learnable decoder, while completely omitting a learnable encoder. Instead, LessNet replaces the encoder with simple, handcrafted features, eliminating the need to optimize encoder parameters. This results in a compact, efficient, and decoder-only architecture for 3D medical image registration. We evaluate our decoder-only LessNet on five registration tasks: 1) inter-subject brain registration using the OASIS-1 dataset, 2) atlas-based brain registration using the IXI dataset, 3) cardiac ES-ED registration using the ACDC dataset, 4) inter-subject abdominal MR registration using the CHAOS dataset, and 5) multi-study, multi-site brain registration using images from 13 public datasets. Our results demonstrate that LessNet can effectively and efficiently learn both dense displacement and diffeomorphic deformation fields. Furthermore, our decoder-only LessNet can achieve comparable registration performance to benchmarking methods such as VoxelMorph and TransMorph, while requiring significantly fewer computational resources. Our code and pre-trained models are available at https://github.com/xi-jia/LessNet
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 23
ISSN:0278-0062
1558-254X
1558-254X
DOI:10.1109/TMI.2025.3562056