View in EDS

Robust multi-domain digital pathology image segmentation via joint balancing representation learning.

Saved in:
Bibliographic Details
Title: Robust multi-domain digital pathology image segmentation via joint balancing representation learning.
Authors: Xu, Qiaoyi1 (AUTHOR) p114345@siswa.ukm.edu.my, Adam, Afzan1,2 (AUTHOR) afzan@ukm.edu.my, Abdullah, Azizi1 (AUTHOR) azizia@ukm.edu.my, Chen, Tao3 (AUTHOR) t66chen@uwaterloo.ca, Zhang, Xinglin4 (AUTHOR) xinglinzhang@imagecore.com.cn, Shephard, Adam2 (AUTHOR) Adam.Shephard@warwick.ac.uk, Pei Sze, Patsy Ng5 (AUTHOR) peisze.ng@premierintegratedlabs.com.my, Masir, Noraidah6 (AUTHOR) noraidah.masir@premierintegratedlabs.com.my, Li, Lei1,7 (AUTHOR) lenny.lilei.cs@gmail.com, Rahayu, Reena8 (AUTHOR) reenarahayu@ppukm.ukm.edu.my
Source: Expert Systems with Applications. Jul2026, Vol. 320, pN.PAG-N.PAG. 1p.
Subjects: Image segmentation, Machine learning, Breast tumors, Generalization, Tissue analysis, Digital diagnostic imaging
Abstract: • Domain contribution imbalance is identified as a key challenge in multi-domain learning. • A joint balancing representation learning framework is proposed to address this issue. • Boundary generalization is explicitly modeled via cross-domain consistency. • Robust segmentation is achieved across heterogeneous pathology domains. [Display omitted] Multi-domain learning (MDL) seeks to mitigate performance degradation caused by domain-specific feature shifts between training and testing environments. In breast cancer digital pathology, such shifts result from variations in staining protocols, imaging devices, and tissue preparation. Existing MDL methods focus on aligning feature discrepancies across domains, often neglecting inter-domain feature balance caused by data distribution disparities. Addressing this balance is essential for effective cross-domain learning in breast cancer pathology, particularly in clinically diverse settings. Additionally, leveraging multi-source training data to enhance model adaptability across pathological domains remains challenging. We introduce a Joint Training Strategy (JTS) and a novel breast cancer digital pathology dataset with expert annotations to capture pathological heterogeneity across multiple domains. We propose Differential Ratio Integration with Twin-domain Training (DRIFT) for multi-source digital pathology image segmentation, addressing domain adaptation through: (1) PRISM-DD, a dynamic data distribution mechanism that balances domain contributions to optimize segmentation, and (2) AMBiCoL, a multi-level loss function integrating adaptive boundary masks with bidirectional consistency modeling to enhance generalization. Experiments on two breast cancer digital pathology datasets demonstrate that DRIFT outperforms state-of-the-art methods, supporting its potential for robust, scalable multi-domain segmentation in computational pathology. Code is available at https://github.com/Joycecc123/DRIFT. [ABSTRACT FROM AUTHOR]
Copyright of Expert Systems with Applications is the property of Pergamon Press - An Imprint of Elsevier Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Engineering Source
FullText Text:
  Availability: 0
Header DbId: egs
DbLabel: Engineering Source
An: 194522039
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Robust multi-domain digital pathology image segmentation via joint balancing representation learning.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Xu%2C+Qiaoyi%22">Xu, Qiaoyi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> p114345@siswa.ukm.edu.my</i><br /><searchLink fieldCode="AR" term="%22Adam%2C+Afzan%22">Adam, Afzan</searchLink><relatesTo>1,2</relatesTo> (AUTHOR)<i> afzan@ukm.edu.my</i><br /><searchLink fieldCode="AR" term="%22Abdullah%2C+Azizi%22">Abdullah, Azizi</searchLink><relatesTo>1</relatesTo> (AUTHOR)<i> azizia@ukm.edu.my</i><br /><searchLink fieldCode="AR" term="%22Chen%2C+Tao%22">Chen, Tao</searchLink><relatesTo>3</relatesTo> (AUTHOR)<i> t66chen@uwaterloo.ca</i><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Xinglin%22">Zhang, Xinglin</searchLink><relatesTo>4</relatesTo> (AUTHOR)<i> xinglinzhang@imagecore.com.cn</i><br /><searchLink fieldCode="AR" term="%22Shephard%2C+Adam%22">Shephard, Adam</searchLink><relatesTo>2</relatesTo> (AUTHOR)<i> Adam.Shephard@warwick.ac.uk</i><br /><searchLink fieldCode="AR" term="%22Pei+Sze%2C+Patsy+Ng%22">Pei Sze, Patsy Ng</searchLink><relatesTo>5</relatesTo> (AUTHOR)<i> peisze.ng@premierintegratedlabs.com.my</i><br /><searchLink fieldCode="AR" term="%22Masir%2C+Noraidah%22">Masir, Noraidah</searchLink><relatesTo>6</relatesTo> (AUTHOR)<i> noraidah.masir@premierintegratedlabs.com.my</i><br /><searchLink fieldCode="AR" term="%22Li%2C+Lei%22">Li, Lei</searchLink><relatesTo>1,7</relatesTo> (AUTHOR)<i> lenny.lilei.cs@gmail.com</i><br /><searchLink fieldCode="AR" term="%22Rahayu%2C+Reena%22">Rahayu, Reena</searchLink><relatesTo>8</relatesTo> (AUTHOR)<i> reenarahayu@ppukm.ukm.edu.my</i>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Expert+Systems+with+Applications%22">Expert Systems with Applications</searchLink>. Jul2026, Vol. 320, pN.PAG-N.PAG. 1p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Image+segmentation%22">Image segmentation</searchLink><br /><searchLink fieldCode="DE" term="%22Machine+learning%22">Machine learning</searchLink><br /><searchLink fieldCode="DE" term="%22Breast+tumors%22">Breast tumors</searchLink><br /><searchLink fieldCode="DE" term="%22Generalization%22">Generalization</searchLink><br /><searchLink fieldCode="DE" term="%22Tissue+analysis%22">Tissue analysis</searchLink><br /><searchLink fieldCode="DE" term="%22Digital+diagnostic+imaging%22">Digital diagnostic imaging</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: • Domain contribution imbalance is identified as a key challenge in multi-domain learning. • A joint balancing representation learning framework is proposed to address this issue. • Boundary generalization is explicitly modeled via cross-domain consistency. • Robust segmentation is achieved across heterogeneous pathology domains. [Display omitted] Multi-domain learning (MDL) seeks to mitigate performance degradation caused by domain-specific feature shifts between training and testing environments. In breast cancer digital pathology, such shifts result from variations in staining protocols, imaging devices, and tissue preparation. Existing MDL methods focus on aligning feature discrepancies across domains, often neglecting inter-domain feature balance caused by data distribution disparities. Addressing this balance is essential for effective cross-domain learning in breast cancer pathology, particularly in clinically diverse settings. Additionally, leveraging multi-source training data to enhance model adaptability across pathological domains remains challenging. We introduce a Joint Training Strategy (JTS) and a novel breast cancer digital pathology dataset with expert annotations to capture pathological heterogeneity across multiple domains. We propose Differential Ratio Integration with Twin-domain Training (DRIFT) for multi-source digital pathology image segmentation, addressing domain adaptation through: (1) PRISM-DD, a dynamic data distribution mechanism that balances domain contributions to optimize segmentation, and (2) AMBiCoL, a multi-level loss function integrating adaptive boundary masks with bidirectional consistency modeling to enhance generalization. Experiments on two breast cancer digital pathology datasets demonstrate that DRIFT outperforms state-of-the-art methods, supporting its potential for robust, scalable multi-domain segmentation in computational pathology. Code is available at https://github.com/Joycecc123/DRIFT. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Expert Systems with Applications is the property of Pergamon Press - An Imprint of Elsevier Science and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=egs&AN=194522039
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1016/j.eswa.2026.132093
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 1
        StartPage: N.PAG
    Subjects:
      – SubjectFull: Image segmentation
        Type: general
      – SubjectFull: Machine learning
        Type: general
      – SubjectFull: Breast tumors
        Type: general
      – SubjectFull: Generalization
        Type: general
      – SubjectFull: Tissue analysis
        Type: general
      – SubjectFull: Digital diagnostic imaging
        Type: general
    Titles:
      – TitleFull: Robust multi-domain digital pathology image segmentation via joint balancing representation learning.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Xu, Qiaoyi
      – PersonEntity:
          Name:
            NameFull: Adam, Afzan
      – PersonEntity:
          Name:
            NameFull: Abdullah, Azizi
      – PersonEntity:
          Name:
            NameFull: Chen, Tao
      – PersonEntity:
          Name:
            NameFull: Zhang, Xinglin
      – PersonEntity:
          Name:
            NameFull: Shephard, Adam
      – PersonEntity:
          Name:
            NameFull: Pei Sze, Patsy Ng
      – PersonEntity:
          Name:
            NameFull: Masir, Noraidah
      – PersonEntity:
          Name:
            NameFull: Li, Lei
      – PersonEntity:
          Name:
            NameFull: Rahayu, Reena
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 15
              M: 07
              Text: Jul2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 09574174
          Numbering:
            – Type: volume
              Value: 320
          Titles:
            – TitleFull: Expert Systems with Applications
              Type: main
ResultId 1