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

Direct link
Towards Automatic Quantification of Immunohistochemistry Using Colour Image Analysis
Uppsala University, Disciplinary Domain of Science and Technology, Mathematics and Computer Science, Centre for Image Analysis.
1998 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Quantification of the proportions of specifically stained regions in images is of significant interest in a growing number of biomedical applications. These applications includes histology and cytology where quantification of various stainings performed on histological tissue sections, smears, imprints etc. is of utmost importance. Through the use of special stains biological components of interest can be given a specific colour. Qualitatively this can be evaluated visually as the presence of a specific colour. But to perform a quantitative evaluation the number of stained cell nuclei and/or the proportion of specimen area that has been stained needs to be measured. Pure visual estimates of this provide very crude results with poor inter- and intraobserver reproducibility. For this purpose computerised image analysis based methods are needed. The methods presented in this thesis aim to make the quantification objective and reproducible.

A new supervised method for computing a pixelwise box classifier has been developed. The resulting classifier can be applied to images of the same type as the training image as long as the lighting conditions have not been changed. The main advantage of this method is that time will be saved if there are many similar images to classify, since box/classification is a fast method.

In order to reduce user interaction, automatic methods for classification, based on more specific knowledge about the images, were developed. These methods include automatic classification of two types of roundish objects, e.g. cell nuclei, on a lighter background, first without, and then with the help of reference images of external cultured cells stained together with the specimen. A method for automatic segmentation of dark thin structures, e.g. microvessels, has been developed as well.

A characteristic of all these methods is that they are implemented as a sequence of single colour band operations, instead of multiband operations. The purpose of this is to make the operations simple and efficient.

Place, publisher, year, edition, pages
Uppsala: Acta Universitatis Upsaliensis, 1998. , 40 p.
Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology, ISSN 1104-232X ; 347
National Category
Medical Image Processing
Research subject
Computerized Image Analysis
URN: urn:nbn:se:uu:diva-862ISBN: 91-554-4152-1OAI: oai:DiVA.org:uu-862DiVA: diva2:171757
Public defence
1998-03-27, Room 4101, Ångström Laboratory, Lägerhyddsvägen 1, Uppsala, 10:15 (English)
Available from: 1998-03-06 Created: 1998-03-06 Last updated: 2015-01-29Bibliographically approved

Open Access in DiVA

No full text
Buy this publication >>

Search in DiVA

By author/editor
Ranefall, Petter
By organisation
Centre for Image Analysis
Medical Image Processing

Search outside of DiVA

GoogleGoogle Scholar
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

Total: 873 hits
ReferencesLink to record
Permanent link

Direct link