Show simple item record

Files in this item


Item metadata

dc.contributor.authorMorrison, David
dc.contributor.authorHarris-Birtill, David Cameron Christopher
dc.contributor.editorTomaszewski, John E.
dc.contributor.editorWard, Aaron D.
dc.contributor.editorRichard M.
dc.identifier.citationMorrison , D & Harris-Birtill , D C C 2022 , Anonymising pathology data using generative adversarial networks . in J E Tomaszewski , A D Ward & R M (eds) , Medical Imaging 2022 : Digital and Computational Pathology . , 1203917 , Proceedings of SPIE , vol. 12039 , SPIE , SPIE Medical Imaging 2022 , San Diego , California , United States , 20/02/22 .
dc.identifier.otherPURE: 279514272
dc.identifier.otherPURE UUID: 166be641-c36d-4bf1-9075-a9ff48e84589
dc.identifier.otherORCID: /0000-0002-0740-3668/work/112711465
dc.identifier.otherScopus: 85132842323
dc.identifier.otherWOS: 000838055900035
dc.description.abstractAnonymising medical data for use in machine learning is important to preserve patient privacy and, in many circumstances, is a requirement before data can be made available. One approach to anonymising image data is to train a generative model to produce data that is statistically similar to the input data and then use the output of the model for downstream tasks, such as image classification, instead of the original sensitive data. In digital pathology, it's not yet well understood how using generative models to anonymise histology slide data impacts the performance of downstream tasks. To begin addressing this, we present an evaluation of a histology image classifier trained using patches extracted from the Camelyon 16 dataset and compare it to a classifier trained on the same number of synthetic images generated with a Deep Convolutional Generative Adversarial Network (DCGAN), from the same data. When predicting the class of an image patch as either cancer or normal it's shown that the accuracy reduces from 0.78 for original alone to 0.59 for synthetic alone, and the recall is significantly reduced from 0.70 to 0.44 when training exclusively on the same amount of synthetic data. If retaining a similar accuracy is required for the downstream task, then either the original data must be used or an improved anonymisation strategy must be devised. We conclude that using this DCGAN to anonymise the dataset, degrades the accuracy of the classifier which implies that it has failed to capture the required variation in the original data to generalise and act as a sufficient anonymisation strategy.
dc.relation.ispartofMedical Imaging 2022en
dc.relation.ispartofseriesProceedings of SPIEen
dc.rightsCopyright © 2022 Society of Photo-Optical Instrumentation Engineers [SPIE]. This work has been made available online in accordance with publisher policies or with permission. Permission for further reuse of this content should be sought from the publisher or the rights holder. This is the final published version of the work, which was originally published at
dc.subjectGenerative adversarial networksen
dc.subjectDigital pathologyen
dc.subjectMedical anonymisationen
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subjectQA76 Computer softwareen
dc.subjectRB Pathologyen
dc.subjectRC0254 Neoplasms. Tumors. Oncology (including Cancer)en
dc.titleAnonymising pathology data using generative adversarial networksen
dc.typeConference itemen
dc.description.versionPublisher PDFen
dc.description.versionPublisher PDFen
dc.contributor.institutionUniversity of St Andrews. School of Computer Scienceen

This item appears in the following Collection(s)

Show simple item record