Learning Histopathological Patterns

    Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

    1769 Downloads (Pure)

    Abstract

    We propose a technique for analyzing images of immunohistochemically stained tissue samples for extracting features that correlate with patient disease. We address the problem of quantifying tumor tissue and segmenting and counting nuclei. Our method utilizes a flexible segmentation technique trained from representative image samples. Nuclei counting is based on a nucleus model that takes size, shape and nucleus probability into account. We obtain the probability of a nucleus from our segmentation procedure. Our method is experimentally validated on images stained with nuclear markers for the Estrogen Receptor (ER) and proliferation marker KI-67. In addition we qualitatively validate our method for tumor tissue segmentation and we obtain state of the art results on cell nuclei separation.
    Original languageEnglish
    Title of host publicationProceedings of the MICCAI workshop on Histopathology Image Analysis
    Publication date2011
    Publication statusPublished - 2011
    EventMICCAI workshop on Histopathology Image Analysis : Clinical Challenges and Quantitative Image Analysis Solutions - Toronto, Canada
    Duration: 1 Jan 2011 → …

    Conference

    ConferenceMICCAI workshop on Histopathology Image Analysis : Clinical Challenges and Quantitative Image Analysis Solutions
    CityToronto, Canada
    Period01/01/2011 → …

    Fingerprint

    Dive into the research topics of 'Learning Histopathological Patterns'. Together they form a unique fingerprint.

    Cite this