Deep Learning for Histopathological Image Analysis
|Sponsor: TUBITAK - Scientific and Technological Research Council of Turkey|
|Project no: 116E075|
|Principal investigator: Cigdem Gunduz Demir|
|Duration: Apr 2017 - Apr 2019|
Today, histopathological examination is the gold standard for cancer diagnosis. This practice is time consuming and prone to subjectivity as it requires detailed inspection and interpretation of pathologists. Digital pathology aims to mitigate these problems by developing computerized systems on objective measures and facilitating quantitative analysis of digitized images. Their success depends on defining effective features for image representation. Manual definition is challenging and inadequate for many cases. Thus, deep learning models that learn features directly on image data are becoming popular.
The goal of this project is to design and implement new deep learning based classification and segmentation methods for histopathological image analysis.