liu.seSearch for publications in DiVA
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Measuring Domain Shift for Deep Learning in Histopathology
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0003-1066-3070
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering. Linköping University, Center for Medical Image Science and Visualization (CMIV).ORCID iD: 0000-0002-9217-9997
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering.ORCID iD: 0000-0002-7765-1747
Linköping University, Department of Science and Technology, Media and Information Technology. Linköping University, Faculty of Science & Engineering. Linköping University, Center for Medical Image Science and Visualization (CMIV).ORCID iD: 0000-0002-9368-0177
2021 (English)In: IEEE journal of biomedical and health informatics, ISSN 2168-2194, E-ISSN 2168-2208, Vol. 25, no 2, p. 325-336Article in journal (Refereed) Published
Abstract [en]

The high capacity of neural networks allows fitting models to data with high precision, but makes generalization to unseen data a challenge. If a domain shift exists, i.e. differences in image statistics between training and test data, care needs to be taken to ensure reliable deployment in real-world scenarios. In digital pathology, domain shift can be manifested in differences between whole-slide images, introduced by for example differences in acquisition pipeline - between medical centers or over time. In order to harness the great potential presented by deep learning in histopathology, and ensure consistent model behavior, we need a deeper understanding of domain shift and its consequences, such that a model's predictions on new data can be trusted. This work focuses on the internal representation learned by trained convolutional neural networks, and shows how this can be used to formulate a novel measure - the representation shift - for quantifying the magnitude of model specific domain shift. We perform a study on domain shift in tumor classification of hematoxylin and eosin stained images, by considering different datasets, models, and techniques for preparing data in order to reduce the domain shift. The results show how the proposed measure has a high correlation with drop in performance when testing a model across a large number of different types of domain shifts, and how it improves on existing techniques for measuring data shift and uncertainty. The proposed measure can reveal how sensitive a model is to domain variations, and can be used to detect new data that a model will have problems generalizing to. We see techniques for measuring, understanding and overcoming the domain shift as a crucial step towards reliable use of deep learning in the future clinical pathology applications.

Place, publisher, year, edition, pages
IEEE, 2021. Vol. 25, no 2, p. 325-336
Keywords [en]
deep learning, machine learning, domain shift, histopathology
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:liu:diva-170816DOI: 10.1109/JBHI.2020.3032060ISI: 000616310200003OAI: oai:DiVA.org:liu-170816DiVA, id: diva2:1478702
Note

Funding:  Wallenberg AI and Autonomous Systems and Software Program (WASP-AI); research environment ELLIIT; AIDA VinnovaVinnova [2017-02447]

Available from: 2020-10-23 Created: 2020-10-23 Last updated: 2023-04-03
In thesis
1. Deep Learning for Digital Pathology in Limited Data Scenarios
Open this publication in new window or tab >>Deep Learning for Digital Pathology in Limited Data Scenarios
2022 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

The impressive technical advances seen for machine learning algorithms in combination with the digitalization of medical images in the radiology and pathology departments show great promise in introducing powerful image analysis tools for image diagnostics. In particular, deep learning, a subfield within machine learning, has shown great success, advancing fields such as image classification and detection. However, these types of algorithms are only used to a very small extent in clinical practice. 

One reason is that the unique nature of radiology and pathology images and the clinical setting in which they are acquired poses challenges not seen in other image domains. Differences relate to capturing methods, as well as the image contents. In addition, these datasets are not only unique on a per-image basis but as a collective dataset. Characteristics such as size, class balance, and availability of annotated labels make creating robust and generalizable deep learning methods a challenge. 

This thesis investigates how deep learning models can be trained for applications in this domain, with particular focus on histopathology data. We investigate how domain shift between different scanners causes performance drop, and present ways of mitigating this. We also present a method to detect when domain shift occurs between different datasets. Another hurdle is the shortage of labeled data for medical applications, and this thesis looks at two different approaches to solving this problem. The first approach investigates how labeled data from one organ and cancer type can boost cancer classification in another organ where labeled data is scarce. The second approach looks at a specific type of unsupervised learning method, self-supervised learning, where the model is trained on unlabeled data. For both of these approaches, we present strategies to handle low-data regimes that may greatly increase the availability to build deep learning models for a wider range of applications. 

Furthermore, deep learning technology enables us to go beyond traditional medical domains, and combine the data from both radiology and pathology. This thesis presents a method for improved cancer characterization on contrast-enhanced CT by incorporating corresponding pathology data during training. The method shows the potential of im-proving future healthcare by intergraded diagnostics made possible by machine-learning technology. 

Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2022. p. 60
Series
Linköping Studies in Science and Technology. Dissertations, ISSN 0345-7524 ; 2253
Keywords
Medical imaging, Digital pathology, Radiology, Machine learning, Deep learning.
National Category
Cancer and Oncology
Identifiers
urn:nbn:se:liu:diva-189009 (URN)10.3384/9789179294748 (DOI)9789179294731 (ISBN)9789179294748 (ISBN)
Public defence
2022-11-14, Kåkenhus, K3, Campus Norrköping, Norrköping, 09:15 (English)
Opponent
Supervisors
Available from: 2022-10-07 Created: 2022-10-07 Last updated: 2023-04-03Bibliographically approved

Open Access in DiVA

fulltext(11581 kB)6481 downloads
File information
File name FULLTEXT02.pdfFile size 11581 kBChecksum SHA-512
0e30f95595f80cd7c0c1c199c0faa143d2b4d50489ad11e740292d4f717d6096a60797e9a20b9752d7a49bc651cd5470ce1eb9d987a1c291045ab51efa6d344a
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Stacke, KarinEilertsen, GabrielUnger, JonasLundström, Claes

Search in DiVA

By author/editor
Stacke, KarinEilertsen, GabrielUnger, JonasLundström, Claes
By organisation
Media and Information TechnologyFaculty of Science & EngineeringCenter for Medical Image Science and Visualization (CMIV)
In the same journal
IEEE journal of biomedical and health informatics
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 6483 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 4337 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • oxford
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf