uu.seUppsala University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • 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
Visualization of convolutional neural network class activations in automated oral cancer detection for interpretation of malignancy associated changes
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Visual Information and Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Computerized Image Analysis and Human-Computer Interaction. (MIDA)ORCID iD: 0000-0002-2891-5435
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Visual Information and Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Computerized Image Analysis and Human-Computer Interaction.ORCID iD: 0000-0002-6041-6310
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Division of Visual Information and Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Computerized Image Analysis and Human-Computer Interaction. Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Department of Information Technology, Automatic control.ORCID iD: 0000-0002-1636-3469
Pathology and Cytology, Karolinska Institute, Stockholm, Sweden.
Show others and affiliations
2019 (English)In: 3rd NEUBIAS Conference, Luxembourg, 2-8 February 2019, 2019, , p. 1Conference paper, Poster (with or without abstract) (Other academic)
Resource type
Text
Abstract [en]

Introduction: Cancer of the oral cavity is one of the most common malignancies in the world. The incidence of oral cavity and oropharyngeal cancer is increasing among young people. It is noteworthy that the oral cavity can be relatively easily accessed for routine screening tests that could potentially decrease the incidence of oral cancer. Automated deep learning computer aided methods show promising ability for detection of subtle precancerous changes at a very early stage, also when visual examination is less effective. Although the biological nature of these malignancy associated changes is not fully understood, the consistency of morphology and textural changes within a cell dataset could shed light on the premalignant state. In this study, we are aiming to increase understanding of this phenomenon by exploring and visualizing what parts of cell images are considered as most important when trained deep convolutional neural networks (DCNNs) are used to differentiate cytological images into normal and abnormal classes.

Materials and methods: Cell samples are collected with a brush at areas of interest in the oral cavity and stained according to standard PAP procedures. Digital images from the slides are acquired with a 0.32 micron pixel size in greyscale format (570 nm bandpass filter). Cell nuclei are manually selected in the images and a small region is cropped around each nucleus resulting in images of 80x80 pixels. Medical knowledge is not used for choosing the cells but they are just randomly selected from the glass; for the learning process we are only providing ground truth on the patient level and not on the cell level. Overall, 10274 images of cell nuclei and the surrounding region are used to train state-of-the-art DCNNs to distinguish between cells from healthy persons and persons with precancerous lesions. Data augmentation through 90 degrees rotations and mirroring is applied to the datasets. Different approaches for class activation mapping and related methods are utilized to determine what image regions and feature maps are responsible for the relevant class differentiation.

Results and Discussion:The best performing of the observed deep learning architectures reaches a per cell classification accuracy surpassing 80% on the observed material. Visualizing the class activation maps confirms our expectation that the network is able to learn to focus on specific relevant parts of the sample regions. We compare and evaluate our findings related to detected discriminative regions with the subjective judgements of a trained cytotechnologist. We believe that this effort on improving understanding of decision criteria used by machine and human leads to increased understanding of malignancy associated changes and also improves robustness and reliability of the automated malignancy detection procedure.

Place, publisher, year, edition, pages
2019. , p. 1
Keywords [en]
Oral cancer, saliency methods, deep convolutional neural networks
National Category
Other Computer and Information Science Medical Image Processing
Research subject
Computerized Image Processing
Identifiers
URN: urn:nbn:se:uu:diva-398145OAI: oai:DiVA.org:uu-398145DiVA, id: diva2:1374747
Conference
3rd NEUBIAS Conference, Luxembourg, 2-8 February 2019
Funder
Swedish Research Council, 2017-04385Available from: 2019-12-02 Created: 2019-12-02 Last updated: 2019-12-04

Open Access in DiVA

No full text in DiVA

Authority records BETA

Koriakina, NadezhdaSladoje, NatasaBengtsson, EwertHirsch, Jan M.Lindblad, Joakim

Search in DiVA

By author/editor
Koriakina, NadezhdaSladoje, NatasaBengtsson, EwertHirsch, Jan M.Lindblad, Joakim
By organisation
Division of Visual Information and InteractionComputerized Image Analysis and Human-Computer InteractionAutomatic controlOral and Maxillofacial Surgery
Other Computer and Information ScienceMedical Image Processing

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 32 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association
  • vancouver
  • 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