Show simple item record

dc.contributor.advisorEngan, Kjersti
dc.contributor.advisorEftestøl, Trygve
dc.contributor.authorWetteland, Rune
dc.date.accessioned2022-02-07T12:36:14Z
dc.date.available2022-02-07T12:36:14Z
dc.date.issued2022-02
dc.identifier.citationAutomated Grading of Bladder Cancer using Deep Learning by Rune Wetteland, Stavanger : University of Stavanger, 2022 (PhD thesis UiS, no. 624)en_US
dc.identifier.isbn978-82-8439-056-7
dc.identifier.issn1890-1387
dc.identifier.urihttps://hdl.handle.net/11250/2977487
dc.descriptionPhD thesis in Information technologyen_US
dc.description.abstractUrothelial carcinoma is the most common type of bladder cancer and is among the cancer types with the highest recurrence rate and lifetime treatment cost per patient. Diagnosed patients are stratified into risk groups, mainly based on the histological grade and stage. However, it is well known that correct grading of bladder cancer suffers from intra- and interobserver variability and inconsistent reproducibility between pathologists, potentially leading to under- or overtreatment of the patients. The economic burden, unnecessary patient suffering, and additional load on the health care system illustrate the importance of developing new tools to aid pathologists. With the introduction of digital pathology, large amounts of data have been made available in the form of digital histological whole-slide images (WSI). However, despite the massive amount of data, annotations for the given data are lacking. Another potential problem is that the tissue samples of urothelial carcinoma contain a mixture of damaged tissue, blood, stroma, muscle, and urothelium, where it is mainly the urothelium tissue that is diagnostically relevant for grading. A method for tissue segmentation is investigated, where the aim is to segment WSIs into the six tissue classes: urothelium, stroma, muscle, damaged tissue, blood, and background. Several methods based on convolutional neural networks (CNN) for tile-wise classification are proposed. Both single-scale and multiscale models were explored to see if including more magnification levels would improve the performance. Different techniques, such as unsupervised learning, semi-supervised learning, and domain adaptation techniques, are explored to mitigate the challenge of missing large quantities of annotated data. It is necessary to extract tiles from the WSI since it is intractable to process the entire WSI at full resolution at once. We have proposed a method to parameterize and automate the task of extracting tiles from different scales with a region of interest (ROI) defined at one of the scales. The method is reproducible and easy to describe by reporting the parameters. A pipeline for automated diagnostic grading is proposed, called TRIgrade. First, the tissue segmentation method is utilized to find the diagnostically relevant urothelium tissue. Then, the parameterized tile extraction method is used to extract tiles from the urothelium regions at three magnification levels from 300 WSIs. The extracted tiles form the training, validation, and test data used to train and test a diagnostic model. The final system outputs a segmented tissue image showing all the tissue regions in the WSI, a WHO grade heatmap indicating low- and high-grade carcinoma regions, and finally, a slide-level WHO grade prediction. The proposed TRIgrade pipeline correctly graded 45 of 50 WSIs, achieving an accuracy of 90%.en_US
dc.language.isoengen_US
dc.publisherUniversity of Stavanger, Norwayen_US
dc.relation.ispartofseriesPhD thesis UiS;
dc.relation.ispartofseries;624
dc.relation.haspartPaper 1: Wetteland, R.; Engan, K.; Eftestøl, T.; Kvikstad, V. and Janssen, E. (2019). Multiclass Tissue Classification of Whole-Slide Histological Images using Convolutional Neural Networks. In Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - ICPRAM, pp, 320-327. DOI: 10.5220/0007253603200327en_US
dc.relation.haspartPaper 2: Wetteland, R.; Engan, K.; Eftestøl, T. (2020) A Multiscale Approach for Whole-Slide Image Segmentation of five Tissue Classes in Urothelial Carcinoma Slides. Technology in Cancer Research & Treatment, 19, DOI: 10.1177/1533033820946787en_US
dc.relation.haspartPaper 3: Dalheim, O.N., Wetteland, R., Kvikstad, V. et al. (2020) Semi-supervised tissue segmentation of histological images. Colour and Visual Computing Symposium 2020, Gjøvik, Norway, September 16-17, 2020. http://ceur-ws.org/Vol-2688/en_US
dc.relation.haspartPaper 4: Wetteland, R.; Engan, K.; Eftestøl, T. (13-15 Sept. 2021) Parameterized Extraction of Tiles in Multilevel Gigapixel Images. 2021 12th International Symposium on Image and Signal Processing and Analysis (ISPA). Zagreb, Croatia, DOI: 10.1109/ISPA52656.2021.9552104. This paper is not in Brage due to copyright restrictions.en_US
dc.relation.haspartPaper 5: Wetteland, R., Kvikstad, V., Eftestøl, T., Tøssebro, E., Lillesand, M., Janssen, E.A.M., Engan, K. (2021) Automatic Diagnostic Tool for Predicting Cancer Grade in Bladder Cancer Patients Using Deep Learning. IEEE Access, 9, pp. 115813 - 115825. DOI: 10.1109/ACCESS.2021.3104724en_US
dc.rightsCopyright the author
dc.rightsAn error occurred on the license name.*
dc.rights.uriAn error occurred getting the license - uri.*
dc.subjectblærekreften_US
dc.subjecturothelial carcinomaen_US
dc.subjectdigital patologien_US
dc.subjectwhole-slide imagesen_US
dc.titleAutomated Grading of Bladder Cancer using Deep Learningen_US
dc.typeDoctoral thesisen_US
dc.rights.holder© 2021 Rune Wettelanden_US
dc.subject.nsiVDP::Teknologi: 500::Medisinsk teknologi: 620en_US
dc.subject.nsiVDP::Medisinske Fag: 700::Klinisk medisinske fag: 750::Onkologi: 762en_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record