Hematology meets Imaging

 

Motivation

Blood cells are formed in the bone marrow. As mature blood cells, they later fulfill numerous important tasks. In diseases such as leukemia, typical patterns change such that fewer cells finish maturing in the bone marrow and more precursors or intermediates are found. Automated recognition of the cell type of a cell and evaluation of an entire bone marrow smear in terms of the number and distribution of certain cell types would accelerate diagnosis, objectify it and open up new avenues in cancer research.

Questions

The central task is to determine the distribution of cell types in the bone marrow. This requires reliable cell detection and classification. However, besides the large number of artifacts and cell types, other typical problems of medical imaging apply to this project: the images are extremely diverse and at the same time the amount of annotated data is limited.

Theses

If you are interested in writing a master’s thesis, I would be pleased to receive a short message. I can then present possible work topics in more detail in a personal meeting!

Partners

The project is carried out in cooperation with the Department of Hematology, Oncology, Hemostaseology and Stem Cell Transplantation of the RWTH University Hospital Aachen. Thus, synergies of medical expertise and digital image processing can be optimally used for innovative research in both fields.

Contact

Publications

2021

Philipp Gräbel, Özcan Özkan, Martina Crysandt, Reinhild Herwartz, Melanie Bauman, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
State of the Art Cell Detection in Bone Marrow Slide Images
In: Journal of Pathology Informatics 12 (1)


2021

Philipp Gräbel, Martina Crysandt, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Guided Representation Learning for the Classification of Hematopoietic Cells
In: ICCV Workshop on Computational Challenges in Digital Pathology (CDPath)


2021

Philipp Gräbel, Ina Laube, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Surrounding Cell Suppression for Unsupervised Representation Learning in Hematological Cell Classification
In: IEEE International Symposium on Biomedical Imaging (ISBI)


2021

Philipp Gräbel, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Reduction of stain variability in bone marrow microscopy images
In: Bildverarbeitung fuer die Medizin (BVM)


2021

Philipp Gräbel, Ina Laube, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Rotation invariance for unsupervised cell representation learning
In: Bildverarbeitung fuer die Medizin (BVM)


2020

Philipp Gräbel, Gregor Nickel, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Systematic Analysis and Automated Search of Hyper-parameters for Cell Classifier Training
In: IEEE International Symposium on Biomedical Imaging (ISBI)


2020

Philipp Gräbel, Özcan Özkan, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Circular Anchors for the Detection of Hematopoietic Cells using RetinaNet
In: IEEE International Symposium on Biomedical Imaging (ISBI)


2019

Philipp Gräbel, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Refinement of Weak Annotations for the Segmentation of Bone Marrow Leukocytes
In: 2nd MICCAI Workshop on Computational Pathology (COMPAY)


2018

Philipp Gräbel, Martina Crysandt, Reinhild Herwartz, Melanie Hoffmann, Barbara M. Klinkhammer, Peter Boor, Tim H. Brümmendorf, Dorit Merhof
Evaluating Out-of-the-box Methods for the Classification of Hematopoietic Cells in Images of Stained Bone Marrow
In: 1st MICCAI Workshop on Computational Pathology (COMPAY)