ARHNet: Adaptive Region Harmonization for Lesion-Aware Augmentation to Improve Segmentation Performance

Jiayu Huo*, Yang Liu, Xi Ouyang, Alejandro Granados, Sébastien Ourselin, Rachel Sparks

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference paperpeer-review

Abstract

Accurately segmenting brain lesions in MRI scans is critical for providing patients with prognoses and neurological monitoring. However, the performance of CNN-based segmentation methods is constrained by the limited training set size. Advanced data augmentation is an effective strategy to improve the model’s robustness. However, they often introduce intensity disparities between foreground and background areas and boundary artifacts, which weakens the effectiveness of such strategies. In this paper, we propose a foreground harmonization framework (ARHNet) to tackle intensity disparities and make synthetic images look more realistic. In particular, we propose an Adaptive Region Harmonization (ARH) module to dynamically align foreground feature maps to the background with an attention mechanism. We demonstrate the efficacy of our method in improving the segmentation performance using real and synthetic images. Experimental results on the ATLAS 2.0 dataset show that ARHNet outperforms other methods for image harmonization tasks, and boosts the down-stream segmentation performance. Our code is publicly available at https://github.com/King-HAW/ARHNet.

Original languageEnglish
Title of host publicationMachine Learning in Medical Imaging - 14th International Workshop, MLMI 2023, Held in Conjunction with MICCAI 2023, Proceedings
EditorsXiaohuan Cao, Xi Ouyang, Xuanang Xu, Islem Rekik, Zhiming Cui
PublisherSpringer Science and Business Media Deutschland GmbH
Pages377-386
Number of pages10
ISBN (Print)9783031456756
DOIs
Publication statusPublished - 2024
Event14th International Workshop on Machine Learning in Medical Imaging, MLMI 2023 - Vancouver, Canada
Duration: 8 Oct 20238 Oct 2023

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume14349 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference14th International Workshop on Machine Learning in Medical Imaging, MLMI 2023
Country/TerritoryCanada
CityVancouver
Period8/10/20238/10/2023

Keywords

  • Adaptive image harmonization
  • Lesion-aware augmentation
  • Stroke segmentation

Fingerprint

Dive into the research topics of 'ARHNet: Adaptive Region Harmonization for Lesion-Aware Augmentation to Improve Segmentation Performance'. Together they form a unique fingerprint.

Cite this