R3Net: Recursive Residual Refinement Network Architecture for Decoder-Free Medical Image Segmentation.

R3Net: Recursive Residual Refinement Network Architecture for Decoder-Free Medical Image Segmentation.

Huang, Jing; Zhao, Yongkang; Li, Yuhan; Dai, Zhitao; Chen, Cheng; Lai, Qiying
Precision radiation oncology 2026
5
jing2026r3net

Abstract

Medical image segmentation methods based on encoder-decoder architectures often achieve high accuracy but typically require substantial computational resources and contain redundant parameters. This study aims to develop an efficient decoder-free segmentation framework that maintains competitive performance by strengthening the encoding process. We propose R3Net, an encoder-only segmentation architecture based on a recursive residual refinement (R3) mechanism. By recursively reusing encoder stages and progressively fusing multiscale features through residual pathways, R3Net reconstructs high-resolution features without requiring a dedicated decoder. Experiments on three medical imaging modalities-cardiac MRI (Automated Cardiac Diagnosis Challenge), abdominal CT (Synapse), and thyroid ultrasound (Thyroid Nodule Multimodal Learning)-demonstrate that R3Net achieves segmentation performance comparable to representative encoder-decoder models while reducing the number of model parameters and computational complexity. R3Net provides an effective decoder-free alternative for medical image segmentation, suggesting that competitive dense prediction can be achieved through recursive refinement within the encoder.

Citation

ID: 284515
Ref Key: jing2026r3net
Use this key to autocite in SciMatic or Thesis Manager

References

Blockchain Verification

Account:
NFT Contract Address:
0x95644003c57E6F55A65596E3D9Eac6813e3566dA
Article ID:
284515
Unique Identifier:
10.1002/pro6.70082
Network:
Scimatic Chain (ID: 481)
Loading...
Blockchain Readiness Checklist
Authors
Abstract
Journal Name
Year
Title
5/5
Creates 1,000,000 NFT tokens for this article
Token Features:
  • ERC-1155 Standard NFT
  • 1 Million Supply per Article
  • Transferable via MetaMask
  • Permanent Blockchain Record
Blockchain QR Code
Scan with Saymatik Web3.0 Wallet

Saymatik Web3.0 Wallet