Segmentation and classification from whole slide images

External Member

David Ahmedt-Aristizabal

Description

Abstract: For the purpose of early diagnosis, proper analysis of histology images is essential. During the diagnosis procedure, the specialist evaluates both overall and local tissue organization via whole-slide and microscopy images. The project aims to provide a unified framework for segmentation and classification in histopathological images. The system comprises an image processing component for handling whole slide images and a model for selecting tissue regions and patches (i.e. extract diagnostically relevant regions.). Then, a prediction score is provided directly only from morphological features (this is a binary problem without any ROI annotations beside the label). Dataset: public datasets (e.g. breast histopathology).

Contact: David Ahmedt-Aristizaba

Updated:  1 June 2019/Responsible Officer:  Dean, CECS/Page Contact:  CECS Marketing