Periodical
View in EDS

Towards a general-purpose foundation model for computational pathology

Bibliographic Details
Title: Towards a general-purpose foundation model for computational pathology
Authors: Chen, Richard J., Ding, Tong, Lu, Ming Y., Williamson, Drew F. K., Jaume, Guillaume, Song, Andrew H., Chen, Bowen, Zhang, Andrew, Shao, Daniel, Shaban, Muhammad, Williams, Mane, Oldenburg, Lukas, Weishaupt, Luca L., Wang, Judy J., Vaidya, Anurag, Le, Long Phi, Gerber, Georg, Sahai, Sharifa, Williams, Walt, Mahmood, Faisal
Source: Nature Medicine; March 2024, Vol. 30 Issue: 3 p850-862, 13p
Abstract: Quantitative evaluation of tissue images is crucial for computational pathology (CPath) tasks, requiring the objective characterization of histopathological entities from whole-slide images (WSIs). The high resolution of WSIs and the variability of morphological features present significant challenges, complicating the large-scale annotation of data for high-performance applications. To address this challenge, current efforts have proposed the use of pretrained image encoders through transfer learning from natural image datasets or self-supervised learning on publicly available histopathology datasets, but have not been extensively developed and evaluated across diverse tissue types at scale. We introduce UNI, a general-purpose self-supervised model for pathology, pretrained using more than 100 million images from over 100,000 diagnostic H&E-stained WSIs (>77 TB of data) across 20 major tissue types. The model was evaluated on 34 representative CPath tasks of varying diagnostic difficulty. In addition to outperforming previous state-of-the-art models, we demonstrate new modeling capabilities in CPath such as resolution-agnostic tissue classification, slide classification using few-shot class prototypes, and disease subtyping generalization in classifying up to 108 cancer types in the OncoTree classification system. UNI advances unsupervised representation learning at scale in CPath in terms of both pretraining data and downstream evaluation, enabling data-efficient artificial intelligence models that can generalize and transfer to a wide range of diagnostically challenging tasks and clinical workflows in anatomic pathology.
Database: Supplemental Index
FullText Links:
  – Type: other
Text:
  Availability: 0
CustomLinks:
  – Url: https://login.libproxy.scu.edu/login?url=http://www.nature.com/openurl?genre=article&title=Nature%20Medicine&volume=30&issue=3&spage=850
    Name: Nature Publishing
    Category: fullText
    Text: Full Text from Nature Publishing
    MouseOverText: Full Text from Nature Publishing
  – Url: https://resolver.ebsco.com/c/xy5jbn/result?sid=EBSCO:edo&genre=article&issn=10788956&ISBN=&volume=30&issue=3&date=20240301&spage=850&pages=850-862&title=Nature Medicine&atitle=Towards%20a%20general-purpose%20foundation%20model%20for%20computational%20pathology&aulast=Chen%2C%20Richard%20J.&id=DOI:10.1038/s41591-024-02857-3
    Name: Full Text Finder (for New FTF UI) (s8985755)
    Category: fullText
    Text: Find It @ SCU Libraries
    MouseOverText: Find It @ SCU Libraries
Header DbId: edo
DbLabel: Supplemental Index
An: ejs65806384
RelevancyScore: 992
AccessLevel: 6
PubType: Periodical
PubTypeId: serialPeriodical
PreciseRelevancyScore: 992.037658691406
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Towards a general-purpose foundation model for computational pathology
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Chen%2C+Richard+J%2E%22">Chen, Richard J.</searchLink><br /><searchLink fieldCode="AR" term="%22Ding%2C+Tong%22">Ding, Tong</searchLink><br /><searchLink fieldCode="AR" term="%22Lu%2C+Ming+Y%2E%22">Lu, Ming Y.</searchLink><br /><searchLink fieldCode="AR" term="%22Williamson%2C+Drew+F%2E+K%2E%22">Williamson, Drew F. K.</searchLink><br /><searchLink fieldCode="AR" term="%22Jaume%2C+Guillaume%22">Jaume, Guillaume</searchLink><br /><searchLink fieldCode="AR" term="%22Song%2C+Andrew+H%2E%22">Song, Andrew H.</searchLink><br /><searchLink fieldCode="AR" term="%22Chen%2C+Bowen%22">Chen, Bowen</searchLink><br /><searchLink fieldCode="AR" term="%22Zhang%2C+Andrew%22">Zhang, Andrew</searchLink><br /><searchLink fieldCode="AR" term="%22Shao%2C+Daniel%22">Shao, Daniel</searchLink><br /><searchLink fieldCode="AR" term="%22Shaban%2C+Muhammad%22">Shaban, Muhammad</searchLink><br /><searchLink fieldCode="AR" term="%22Williams%2C+Mane%22">Williams, Mane</searchLink><br /><searchLink fieldCode="AR" term="%22Oldenburg%2C+Lukas%22">Oldenburg, Lukas</searchLink><br /><searchLink fieldCode="AR" term="%22Weishaupt%2C+Luca+L%2E%22">Weishaupt, Luca L.</searchLink><br /><searchLink fieldCode="AR" term="%22Wang%2C+Judy+J%2E%22">Wang, Judy J.</searchLink><br /><searchLink fieldCode="AR" term="%22Vaidya%2C+Anurag%22">Vaidya, Anurag</searchLink><br /><searchLink fieldCode="AR" term="%22Le%2C+Long+Phi%22">Le, Long Phi</searchLink><br /><searchLink fieldCode="AR" term="%22Gerber%2C+Georg%22">Gerber, Georg</searchLink><br /><searchLink fieldCode="AR" term="%22Sahai%2C+Sharifa%22">Sahai, Sharifa</searchLink><br /><searchLink fieldCode="AR" term="%22Williams%2C+Walt%22">Williams, Walt</searchLink><br /><searchLink fieldCode="AR" term="%22Mahmood%2C+Faisal%22">Mahmood, Faisal</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: Nature Medicine; March 2024, Vol. 30 Issue: 3 p850-862, 13p
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Quantitative evaluation of tissue images is crucial for computational pathology (CPath) tasks, requiring the objective characterization of histopathological entities from whole-slide images (WSIs). The high resolution of WSIs and the variability of morphological features present significant challenges, complicating the large-scale annotation of data for high-performance applications. To address this challenge, current efforts have proposed the use of pretrained image encoders through transfer learning from natural image datasets or self-supervised learning on publicly available histopathology datasets, but have not been extensively developed and evaluated across diverse tissue types at scale. We introduce UNI, a general-purpose self-supervised model for pathology, pretrained using more than 100 million images from over 100,000 diagnostic H&E-stained WSIs (>77 TB of data) across 20 major tissue types. The model was evaluated on 34 representative CPath tasks of varying diagnostic difficulty. In addition to outperforming previous state-of-the-art models, we demonstrate new modeling capabilities in CPath such as resolution-agnostic tissue classification, slide classification using few-shot class prototypes, and disease subtyping generalization in classifying up to 108 cancer types in the OncoTree classification system. UNI advances unsupervised representation learning at scale in CPath in terms of both pretraining data and downstream evaluation, enabling data-efficient artificial intelligence models that can generalize and transfer to a wide range of diagnostically challenging tasks and clinical workflows in anatomic pathology.
PLink https://login.libproxy.scu.edu/login?url=https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&scope=site&db=edo&AN=ejs65806384
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1038/s41591-024-02857-3
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 13
        StartPage: 850
    Titles:
      – TitleFull: Towards a general-purpose foundation model for computational pathology
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Chen, Richard J.
      – PersonEntity:
          Name:
            NameFull: Ding, Tong
      – PersonEntity:
          Name:
            NameFull: Lu, Ming Y.
      – PersonEntity:
          Name:
            NameFull: Williamson, Drew F. K.
      – PersonEntity:
          Name:
            NameFull: Jaume, Guillaume
      – PersonEntity:
          Name:
            NameFull: Song, Andrew H.
      – PersonEntity:
          Name:
            NameFull: Chen, Bowen
      – PersonEntity:
          Name:
            NameFull: Zhang, Andrew
      – PersonEntity:
          Name:
            NameFull: Shao, Daniel
      – PersonEntity:
          Name:
            NameFull: Shaban, Muhammad
      – PersonEntity:
          Name:
            NameFull: Williams, Mane
      – PersonEntity:
          Name:
            NameFull: Oldenburg, Lukas
      – PersonEntity:
          Name:
            NameFull: Weishaupt, Luca L.
      – PersonEntity:
          Name:
            NameFull: Wang, Judy J.
      – PersonEntity:
          Name:
            NameFull: Vaidya, Anurag
      – PersonEntity:
          Name:
            NameFull: Le, Long Phi
      – PersonEntity:
          Name:
            NameFull: Gerber, Georg
      – PersonEntity:
          Name:
            NameFull: Sahai, Sharifa
      – PersonEntity:
          Name:
            NameFull: Williams, Walt
      – PersonEntity:
          Name:
            NameFull: Mahmood, Faisal
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 03
              Text: March 2024
              Type: published
              Y: 2024
          Identifiers:
            – Type: issn-print
              Value: 10788956
            – Type: issn-print
              Value: 1546170X
          Numbering:
            – Type: volume
              Value: 30
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Nature Medicine
              Type: main
ResultId 1