Zero-shot Bias Correction: Efficient MR Image Inhomogeneity Reduction Without Any Data

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Hongxu, Timko, Edina, Fernandez, Brice
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908832635027456
author Yang, Hongxu
Timko, Edina
Fernandez, Brice
author_facet Yang, Hongxu
Timko, Edina
Fernandez, Brice
contents In recent years, deep neural networks for image inhomogeneity reduction have shown promising results. However, current methods with (un)supervised solutions require preparing a training dataset, which is expensive and laborious for data collection. In this work, we demonstrate a novel zero-shot deep neural networks, which requires no data for pre-training and dedicated assumption of the bias field. The designed light-weight CNN enables an efficient zero-shot adaptation for bias-corrupted image correction. Our method provides a novel solution to mitigate the biased corrupted image as iterative homogeneity refinement, which therefore ensures the considered issue can be solved easier with stable convergence of zero-shot optimization. Extensive comparison on different datasets show that the proposed method performs better than current data-free N4 methods in both efficiency and accuracy.
format Preprint
id arxiv_https___arxiv_org_abs_2501_12244
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Zero-shot Bias Correction: Efficient MR Image Inhomogeneity Reduction Without Any Data
Yang, Hongxu
Timko, Edina
Fernandez, Brice
Image and Video Processing
Computer Vision and Pattern Recognition
In recent years, deep neural networks for image inhomogeneity reduction have shown promising results. However, current methods with (un)supervised solutions require preparing a training dataset, which is expensive and laborious for data collection. In this work, we demonstrate a novel zero-shot deep neural networks, which requires no data for pre-training and dedicated assumption of the bias field. The designed light-weight CNN enables an efficient zero-shot adaptation for bias-corrupted image correction. Our method provides a novel solution to mitigate the biased corrupted image as iterative homogeneity refinement, which therefore ensures the considered issue can be solved easier with stable convergence of zero-shot optimization. Extensive comparison on different datasets show that the proposed method performs better than current data-free N4 methods in both efficiency and accuracy.
title Zero-shot Bias Correction: Efficient MR Image Inhomogeneity Reduction Without Any Data
topic Image and Video Processing
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2501.12244