Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.

Multi-source domain adaptation leverages complementary knowledge from multiple source domains to enhance transfer effectiveness, making it more suitable for complex medical scenarios compared to single-source domain adaptation. However, most existing studies operate under the assumption that the sou...

Full description

Saved in:
Bibliographic Details
Main Authors: Huiying Zhang, Yongmeng Li, Lei He, Wenbo Zhang, Yuchen Shen, Lumin Xing
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-01-01
Series:PLoS ONE
Online Access:https://doi.org/10.1371/journal.pone.0323676
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849248080280420352
author Huiying Zhang
Yongmeng Li
Lei He
Wenbo Zhang
Yuchen Shen
Lumin Xing
author_facet Huiying Zhang
Yongmeng Li
Lei He
Wenbo Zhang
Yuchen Shen
Lumin Xing
author_sort Huiying Zhang
collection DOAJ
description Multi-source domain adaptation leverages complementary knowledge from multiple source domains to enhance transfer effectiveness, making it more suitable for complex medical scenarios compared to single-source domain adaptation. However, most existing studies operate under the assumption that the source and target domains share identical class distributions, leaving the challenge of addressing class shift in multi-source domain adaptation largely unexplored. To address this gap, this study proposes a Class-Aware Multi-Source Domain Adaptation algorithm based on a Reweighted Matrix Matching strategy (CAMSDA-RMM). This algorithm employs a class-aware strategy to strengthen positive transfer effects between similar classes. Additionally, first-order and second-order moment matching strategies are applied to effectively align the source and target domains, while an adaptive weighting mechanism is utilized to optimize the contributions of different source domains to the target domain. These approaches collectively improve classification accuracy and domain adaptability. Experimental results on four publicly available chest X-ray datasets demonstrate that the superiority of the proposed method.
format Article
id doaj-art-a8a6906c40374182a5d788dacb2e7d3d
institution Kabale University
issn 1932-6203
language English
publishDate 2025-01-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS ONE
spelling doaj-art-a8a6906c40374182a5d788dacb2e7d3d2025-08-20T03:58:03ZengPublic Library of Science (PLoS)PLoS ONE1932-62032025-01-01207e032367610.1371/journal.pone.0323676Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.Huiying ZhangYongmeng LiLei HeWenbo ZhangYuchen ShenLumin XingMulti-source domain adaptation leverages complementary knowledge from multiple source domains to enhance transfer effectiveness, making it more suitable for complex medical scenarios compared to single-source domain adaptation. However, most existing studies operate under the assumption that the source and target domains share identical class distributions, leaving the challenge of addressing class shift in multi-source domain adaptation largely unexplored. To address this gap, this study proposes a Class-Aware Multi-Source Domain Adaptation algorithm based on a Reweighted Matrix Matching strategy (CAMSDA-RMM). This algorithm employs a class-aware strategy to strengthen positive transfer effects between similar classes. Additionally, first-order and second-order moment matching strategies are applied to effectively align the source and target domains, while an adaptive weighting mechanism is utilized to optimize the contributions of different source domains to the target domain. These approaches collectively improve classification accuracy and domain adaptability. Experimental results on four publicly available chest X-ray datasets demonstrate that the superiority of the proposed method.https://doi.org/10.1371/journal.pone.0323676
spellingShingle Huiying Zhang
Yongmeng Li
Lei He
Wenbo Zhang
Yuchen Shen
Lumin Xing
Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
PLoS ONE
title Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
title_full Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
title_fullStr Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
title_full_unstemmed Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
title_short Class-aware multi-source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy.
title_sort class aware multi source domain adaptation algorithm for medical image analysis using reweighted matrix matching strategy
url https://doi.org/10.1371/journal.pone.0323676
work_keys_str_mv AT huiyingzhang classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy
AT yongmengli classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy
AT leihe classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy
AT wenbozhang classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy
AT yuchenshen classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy
AT luminxing classawaremultisourcedomainadaptationalgorithmformedicalimageanalysisusingreweightedmatrixmatchingstrategy